id
large_stringlengths 9
16
| submitter
large_stringlengths 3
64
⌀ | authors
large_stringlengths 5
20.2k
| title
large_stringlengths 5
245
| comments
large_stringlengths 1
482
⌀ | journal-ref
large_stringlengths 4
382
⌀ | doi
large_stringlengths 9
151
⌀ | report-no
large_stringlengths 2
509
⌀ | categories
large_stringlengths 5
78
| license
large_stringclasses 9
values | abstract
large_stringlengths 83
3.55k
| versions
large listlengths 1
21
| update_date
timestamp[ms]date 2007-05-23 00:00:00
2025-05-30 00:00:00
| authors_parsed
large listlengths 1
1.35k
| prompt
large_stringlengths 166
3.64k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2112.07337 | Saneem Ahmed Chemmengath | Vishwajeet Kumar, Yash Gupta, Saneem Chemmengath, Jaydeep Sen, Soumen
Chakrabarti, Samarth Bharadwaj, FeiFei Pan | Multi-Row, Multi-Span Distant Supervision For Table+Text Question | ACL 2023 | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Question answering (QA) over tables and linked text, also called TextTableQA,
has witnessed significant research in recent years, as tables are often found
embedded in documents along with related text. HybridQA and OTT-QA are the two
best-known TextTableQA datasets, with questions that are best answered by
combining information from both table cells and linked text passages. A common
challenge in both datasets, and TextTableQA in general, is that the training
instances include just the question and answer, where the gold answer may match
not only multiple table cells across table rows but also multiple text spans
within the scope of a table row and its associated text. This leads to a noisy
multi instance training regime. We present MITQA, a transformer-based
TextTableQA system that is explicitly designed to cope with distant supervision
along both these axes, through a multi-instance loss objective, together with
careful curriculum design. Our experiments show that the proposed
multi-instance distant supervision approach helps MITQA get state-of-the-art
results beating the existing baselines for both HybridQA and OTT-QA, putting
MITQA at the top of HybridQA leaderboard with best EM and F1 scores on a held
out test set.
| [
{
"version": "v1",
"created": "Tue, 14 Dec 2021 12:48:19 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Oct 2022 15:26:50 GMT"
},
{
"version": "v3",
"created": "Sun, 11 Jun 2023 18:46:16 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Kumar",
"Vishwajeet",
""
],
[
"Gupta",
"Yash",
""
],
[
"Chemmengath",
"Saneem",
""
],
[
"Sen",
"Jaydeep",
""
],
[
"Chakrabarti",
"Soumen",
""
],
[
"Bharadwaj",
"Samarth",
""
],
[
"Pan",
"FeiFei",
""
]
] | TITLE: Multi-Row, Multi-Span Distant Supervision For Table+Text Question
ABSTRACT: Question answering (QA) over tables and linked text, also called TextTableQA,
has witnessed significant research in recent years, as tables are often found
embedded in documents along with related text. HybridQA and OTT-QA are the two
best-known TextTableQA datasets, with questions that are best answered by
combining information from both table cells and linked text passages. A common
challenge in both datasets, and TextTableQA in general, is that the training
instances include just the question and answer, where the gold answer may match
not only multiple table cells across table rows but also multiple text spans
within the scope of a table row and its associated text. This leads to a noisy
multi instance training regime. We present MITQA, a transformer-based
TextTableQA system that is explicitly designed to cope with distant supervision
along both these axes, through a multi-instance loss objective, together with
careful curriculum design. Our experiments show that the proposed
multi-instance distant supervision approach helps MITQA get state-of-the-art
results beating the existing baselines for both HybridQA and OTT-QA, putting
MITQA at the top of HybridQA leaderboard with best EM and F1 scores on a held
out test set.
|
2112.13002 | Arbish Akram | Arbish Akram and Nazar Khan | US-GAN: On the importance of Ultimate Skip Connection for Facial
Expression Synthesis | null | Multimed Tools Appl (2023) | 10.1007/s11042-023-15268-2 | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We demonstrate the benefit of using an ultimate skip (US) connection for
facial expression synthesis using generative adversarial networks (GAN). A
direct connection transfers identity, facial, and color details from input to
output while suppressing artifacts. The intermediate layers can therefore focus
on expression generation only. This leads to a light-weight US-GAN model
comprised of encoding layers, a single residual block, decoding layers, and an
ultimate skip connection from input to output. US-GAN has $3\times$ fewer
parameters than state-of-the-art models and is trained on $2$ orders of
magnitude smaller dataset. It yields $7\%$ increase in face verification score
(FVS) and $27\%$ decrease in average content distance (ACD). Based on a
randomized user-study, US-GAN outperforms the state of the art by $25\%$ in
face realism, $43\%$ in expression quality, and $58\%$ in identity
preservation.
| [
{
"version": "v1",
"created": "Fri, 24 Dec 2021 08:56:50 GMT"
},
{
"version": "v2",
"created": "Sat, 8 Apr 2023 01:42:24 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Akram",
"Arbish",
""
],
[
"Khan",
"Nazar",
""
]
] | TITLE: US-GAN: On the importance of Ultimate Skip Connection for Facial
Expression Synthesis
ABSTRACT: We demonstrate the benefit of using an ultimate skip (US) connection for
facial expression synthesis using generative adversarial networks (GAN). A
direct connection transfers identity, facial, and color details from input to
output while suppressing artifacts. The intermediate layers can therefore focus
on expression generation only. This leads to a light-weight US-GAN model
comprised of encoding layers, a single residual block, decoding layers, and an
ultimate skip connection from input to output. US-GAN has $3\times$ fewer
parameters than state-of-the-art models and is trained on $2$ orders of
magnitude smaller dataset. It yields $7\%$ increase in face verification score
(FVS) and $27\%$ decrease in average content distance (ACD). Based on a
randomized user-study, US-GAN outperforms the state of the art by $25\%$ in
face realism, $43\%$ in expression quality, and $58\%$ in identity
preservation.
|
2203.01391 | Nail Ibrahimli | Nail Ibrahimli, Hugo Ledoux, Julian Kooij, Liangliang Nan | DDL-MVS: Depth Discontinuity Learning for MVS Networks | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Traditional MVS methods have good accuracy but struggle with completeness,
while recently developed learning-based multi-view stereo (MVS) techniques have
improved completeness except accuracy being compromised. We propose depth
discontinuity learning for MVS methods, which further improves accuracy while
retaining the completeness of the reconstruction. Our idea is to jointly
estimate the depth and boundary maps where the boundary maps are explicitly
used for further refinement of the depth maps. We validate our idea and
demonstrate that our strategies can be easily integrated into the existing
learning-based MVS pipeline where the reconstruction depends on high-quality
depth map estimation. Extensive experiments on various datasets show that our
method improves reconstruction quality compared to baseline. Experiments also
demonstrate that the presented model and strategies have good generalization
capabilities. The source code will be available soon.
| [
{
"version": "v1",
"created": "Wed, 2 Mar 2022 20:25:31 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Mar 2022 11:40:59 GMT"
},
{
"version": "v3",
"created": "Mon, 12 Jun 2023 12:05:42 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Ibrahimli",
"Nail",
""
],
[
"Ledoux",
"Hugo",
""
],
[
"Kooij",
"Julian",
""
],
[
"Nan",
"Liangliang",
""
]
] | TITLE: DDL-MVS: Depth Discontinuity Learning for MVS Networks
ABSTRACT: Traditional MVS methods have good accuracy but struggle with completeness,
while recently developed learning-based multi-view stereo (MVS) techniques have
improved completeness except accuracy being compromised. We propose depth
discontinuity learning for MVS methods, which further improves accuracy while
retaining the completeness of the reconstruction. Our idea is to jointly
estimate the depth and boundary maps where the boundary maps are explicitly
used for further refinement of the depth maps. We validate our idea and
demonstrate that our strategies can be easily integrated into the existing
learning-based MVS pipeline where the reconstruction depends on high-quality
depth map estimation. Extensive experiments on various datasets show that our
method improves reconstruction quality compared to baseline. Experiments also
demonstrate that the presented model and strategies have good generalization
capabilities. The source code will be available soon.
|
2203.14284 | Nir Drucker | Allon Adir, Ehud Aharoni, Nir Drucker, Eyal Kushnir, Ramy Masalha,
Michael Mirkin, Omri Soceanu | Privacy-preserving record linkage using local sensitive hash and private
set intersection | null | null | 10.1007/978-3-031-16815-4_22 | null | cs.CR | http://creativecommons.org/licenses/by/4.0/ | The amount of data stored in data repositories increases every year. This
makes it challenging to link records between different datasets across
companies and even internally, while adhering to privacy regulations. Address
or name changes, and even different spelling used for entity data, can prevent
companies from using private deduplication or record-linking solutions such as
private set intersection (PSI). To this end, we propose a new and efficient
privacy-preserving record linkage (PPRL) protocol that combines PSI and local
sensitive hash (LSH) functions, and runs in linear time. We explain the privacy
guarantees that our protocol provides and demonstrate its practicality by
executing the protocol over two datasets with $2^{20}$ records each, in $11-45$
minutes, depending on network settings.
| [
{
"version": "v1",
"created": "Sun, 27 Mar 2022 11:59:00 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Adir",
"Allon",
""
],
[
"Aharoni",
"Ehud",
""
],
[
"Drucker",
"Nir",
""
],
[
"Kushnir",
"Eyal",
""
],
[
"Masalha",
"Ramy",
""
],
[
"Mirkin",
"Michael",
""
],
[
"Soceanu",
"Omri",
""
]
] | TITLE: Privacy-preserving record linkage using local sensitive hash and private
set intersection
ABSTRACT: The amount of data stored in data repositories increases every year. This
makes it challenging to link records between different datasets across
companies and even internally, while adhering to privacy regulations. Address
or name changes, and even different spelling used for entity data, can prevent
companies from using private deduplication or record-linking solutions such as
private set intersection (PSI). To this end, we propose a new and efficient
privacy-preserving record linkage (PPRL) protocol that combines PSI and local
sensitive hash (LSH) functions, and runs in linear time. We explain the privacy
guarantees that our protocol provides and demonstrate its practicality by
executing the protocol over two datasets with $2^{20}$ records each, in $11-45$
minutes, depending on network settings.
|
2204.01922 | Arec Jamgochian | Arec Jamgochian, Etienne Buehrle, Johannes Fischer, Mykel J.
Kochenderfer | SHAIL: Safety-Aware Hierarchical Adversarial Imitation Learning for
Autonomous Driving in Urban Environments | Presented at the 2023 IEEE International Conference on Robotics and
Automation (ICRA) | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Designing a safe and human-like decision-making system for an autonomous
vehicle is a challenging task. Generative imitation learning is one possible
approach for automating policy-building by leveraging both real-world and
simulated decisions. Previous work that applies generative imitation learning
to autonomous driving policies focuses on learning a low-level controller for
simple settings. However, to scale to complex settings, many autonomous driving
systems combine fixed, safe, optimization-based low-level controllers with
high-level decision-making logic that selects the appropriate task and
associated controller. In this paper, we attempt to bridge this gap in
complexity by employing Safety-Aware Hierarchical Adversarial Imitation
Learning (SHAIL), a method for learning a high-level policy that selects from a
set of low-level controller instances in a way that imitates low-level driving
data on-policy. We introduce an urban roundabout simulator that controls
non-ego vehicles using real data from the Interaction dataset. We then
demonstrate empirically that even with simple controller options, our approach
can produce better behavior than previous approaches in driver imitation that
have difficulty scaling to complex environments. Our implementation is
available at https://github.com/sisl/InteractionImitation.
| [
{
"version": "v1",
"created": "Tue, 5 Apr 2022 01:23:56 GMT"
},
{
"version": "v2",
"created": "Fri, 23 Dec 2022 05:42:39 GMT"
},
{
"version": "v3",
"created": "Sat, 10 Jun 2023 17:56:22 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Jamgochian",
"Arec",
""
],
[
"Buehrle",
"Etienne",
""
],
[
"Fischer",
"Johannes",
""
],
[
"Kochenderfer",
"Mykel J.",
""
]
] | TITLE: SHAIL: Safety-Aware Hierarchical Adversarial Imitation Learning for
Autonomous Driving in Urban Environments
ABSTRACT: Designing a safe and human-like decision-making system for an autonomous
vehicle is a challenging task. Generative imitation learning is one possible
approach for automating policy-building by leveraging both real-world and
simulated decisions. Previous work that applies generative imitation learning
to autonomous driving policies focuses on learning a low-level controller for
simple settings. However, to scale to complex settings, many autonomous driving
systems combine fixed, safe, optimization-based low-level controllers with
high-level decision-making logic that selects the appropriate task and
associated controller. In this paper, we attempt to bridge this gap in
complexity by employing Safety-Aware Hierarchical Adversarial Imitation
Learning (SHAIL), a method for learning a high-level policy that selects from a
set of low-level controller instances in a way that imitates low-level driving
data on-policy. We introduce an urban roundabout simulator that controls
non-ego vehicles using real data from the Interaction dataset. We then
demonstrate empirically that even with simple controller options, our approach
can produce better behavior than previous approaches in driver imitation that
have difficulty scaling to complex environments. Our implementation is
available at https://github.com/sisl/InteractionImitation.
|
2204.02635 | Bo Xu | Bo Xu, Xin Li, Jingrong Wang, Chau Yuen (Fellow, IEEE), Jiancheng Li | PVI-DSO: Leveraging Planar Regularities for Direct Sparse
Visual-Inertial Odometry | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The monocular visual-inertial odometry (VIO) based on the direct method can
leverage all available pixels in the image to simultaneously estimate the
camera motion and reconstruct the denser map of the scene in real time.
However, the direct method is sensitive to photometric changes, which can be
compensated by introducing geometric information in the environment. In this
paper, we propose a monocular direct sparse visual-inertial odometry, which
exploits the planar regularities (PVI-DSO). Our system detects the planar
regularities from the 3D mesh built on the estimated map points. To improve the
pose estimation accuracy with the geometric information, a tightly coupled
coplanar constraint expression is used to express photometric error in the
direct method. Additionally, to improve the optimization efficiency, we
elaborately derive the analytical Jacobian of the linearization form for the
coplanar constraint. Finally, the inertial measurement error, coplanar point
photometric error, non-coplanar photometric error, and prior error are added
into the optimizer, which simultaneously improves the pose estimation accuracy
and mesh itself. We verified the performance of the whole system on simulation
and real-world datasets. Extensive experiments have demonstrated that our
system outperforms the state-of-the-art counterparts.
| [
{
"version": "v1",
"created": "Wed, 6 Apr 2022 07:28:12 GMT"
},
{
"version": "v2",
"created": "Sun, 11 Jun 2023 08:20:30 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Xu",
"Bo",
"",
"Fellow, IEEE"
],
[
"Li",
"Xin",
"",
"Fellow, IEEE"
],
[
"Wang",
"Jingrong",
"",
"Fellow, IEEE"
],
[
"Yuen",
"Chau",
"",
"Fellow, IEEE"
],
[
"Li",
"Jiancheng",
""
]
] | TITLE: PVI-DSO: Leveraging Planar Regularities for Direct Sparse
Visual-Inertial Odometry
ABSTRACT: The monocular visual-inertial odometry (VIO) based on the direct method can
leverage all available pixels in the image to simultaneously estimate the
camera motion and reconstruct the denser map of the scene in real time.
However, the direct method is sensitive to photometric changes, which can be
compensated by introducing geometric information in the environment. In this
paper, we propose a monocular direct sparse visual-inertial odometry, which
exploits the planar regularities (PVI-DSO). Our system detects the planar
regularities from the 3D mesh built on the estimated map points. To improve the
pose estimation accuracy with the geometric information, a tightly coupled
coplanar constraint expression is used to express photometric error in the
direct method. Additionally, to improve the optimization efficiency, we
elaborately derive the analytical Jacobian of the linearization form for the
coplanar constraint. Finally, the inertial measurement error, coplanar point
photometric error, non-coplanar photometric error, and prior error are added
into the optimizer, which simultaneously improves the pose estimation accuracy
and mesh itself. We verified the performance of the whole system on simulation
and real-world datasets. Extensive experiments have demonstrated that our
system outperforms the state-of-the-art counterparts.
|
2204.03444 | Gabriele Berton | Gabriele Berton, Riccardo Mereu, Gabriele Trivigno, Carlo Masone,
Gabriela Csurka, Torsten Sattler, Barbara Caputo | Deep Visual Geo-localization Benchmark | CVPR 2022 (Oral) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose a new open-source benchmarking framework for Visual
Geo-localization (VG) that allows to build, train, and test a wide range of
commonly used architectures, with the flexibility to change individual
components of a geo-localization pipeline. The purpose of this framework is
twofold: i) gaining insights into how different components and design choices
in a VG pipeline impact the final results, both in terms of performance
(recall@N metric) and system requirements (such as execution time and memory
consumption); ii) establish a systematic evaluation protocol for comparing
different methods. Using the proposed framework, we perform a large suite of
experiments which provide criteria for choosing backbone, aggregation and
negative mining depending on the use-case and requirements. We also assess the
impact of engineering techniques like pre/post-processing, data augmentation
and image resizing, showing that better performance can be obtained through
somewhat simple procedures: for example, downscaling the images' resolution to
80% can lead to similar results with a 36% savings in extraction time and
dataset storage requirement. Code and trained models are available at
https://deep-vg-bench.herokuapp.com/.
| [
{
"version": "v1",
"created": "Thu, 7 Apr 2022 13:47:49 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 10:18:20 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Berton",
"Gabriele",
""
],
[
"Mereu",
"Riccardo",
""
],
[
"Trivigno",
"Gabriele",
""
],
[
"Masone",
"Carlo",
""
],
[
"Csurka",
"Gabriela",
""
],
[
"Sattler",
"Torsten",
""
],
[
"Caputo",
"Barbara",
""
]
] | TITLE: Deep Visual Geo-localization Benchmark
ABSTRACT: In this paper, we propose a new open-source benchmarking framework for Visual
Geo-localization (VG) that allows to build, train, and test a wide range of
commonly used architectures, with the flexibility to change individual
components of a geo-localization pipeline. The purpose of this framework is
twofold: i) gaining insights into how different components and design choices
in a VG pipeline impact the final results, both in terms of performance
(recall@N metric) and system requirements (such as execution time and memory
consumption); ii) establish a systematic evaluation protocol for comparing
different methods. Using the proposed framework, we perform a large suite of
experiments which provide criteria for choosing backbone, aggregation and
negative mining depending on the use-case and requirements. We also assess the
impact of engineering techniques like pre/post-processing, data augmentation
and image resizing, showing that better performance can be obtained through
somewhat simple procedures: for example, downscaling the images' resolution to
80% can lead to similar results with a 36% savings in extraction time and
dataset storage requirement. Code and trained models are available at
https://deep-vg-bench.herokuapp.com/.
|
2204.10824 | Ruhui Jin | Ruhui Jin, Joe Kileel, Tamara G. Kolda, Rachel Ward | Scalable symmetric Tucker tensor decomposition | null | null | null | null | math.NA cs.NA | http://creativecommons.org/licenses/by/4.0/ | We study the best low-rank Tucker decomposition of symmetric tensors. The
motivating application is decomposing higher-order multivariate moments. Moment
tensors have special structure and are important to various data science
problems. We advocate for projected gradient descent (PGD) method and
higher-order eigenvalue decomposition (HOEVD) approximation as computation
schemes. Most importantly, we develop scalable adaptations of the basic PGD and
HOEVD methods to decompose sample moment tensors. With the help of implicit and
streaming techniques, we evade the overhead cost of building and storing the
moment tensor. Such reductions make computing the Tucker decomposition
realizable for large data instances in high dimensions. Numerical experiments
demonstrate the efficiency of the algorithms and the applicability of moment
tensor decompositions to real-world datasets. Finally we study the convergence
on the Grassmannian manifold, and prove that the update sequence derived by the
PGD solver achieves first- and second-order criticality.
| [
{
"version": "v1",
"created": "Fri, 22 Apr 2022 17:09:36 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 04:22:46 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Jin",
"Ruhui",
""
],
[
"Kileel",
"Joe",
""
],
[
"Kolda",
"Tamara G.",
""
],
[
"Ward",
"Rachel",
""
]
] | TITLE: Scalable symmetric Tucker tensor decomposition
ABSTRACT: We study the best low-rank Tucker decomposition of symmetric tensors. The
motivating application is decomposing higher-order multivariate moments. Moment
tensors have special structure and are important to various data science
problems. We advocate for projected gradient descent (PGD) method and
higher-order eigenvalue decomposition (HOEVD) approximation as computation
schemes. Most importantly, we develop scalable adaptations of the basic PGD and
HOEVD methods to decompose sample moment tensors. With the help of implicit and
streaming techniques, we evade the overhead cost of building and storing the
moment tensor. Such reductions make computing the Tucker decomposition
realizable for large data instances in high dimensions. Numerical experiments
demonstrate the efficiency of the algorithms and the applicability of moment
tensor decompositions to real-world datasets. Finally we study the convergence
on the Grassmannian manifold, and prove that the update sequence derived by the
PGD solver achieves first- and second-order criticality.
|
2205.12221 | Miyoung Ko | Miyoung Ko, Ingyu Seong, Hwaran Lee, Joonsuk Park, Minsuk Chang,
Minjoon Seo | ClaimDiff: Comparing and Contrasting Claims on Contentious Issues | published at Findings of ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | With the growing importance of detecting misinformation, many studies have
focused on verifying factual claims by retrieving evidence. However, canonical
fact verification tasks do not apply to catching subtle differences in
factually consistent claims, which might still bias the readers, especially on
contentious political or economic issues. Our underlying assumption is that
among the trusted sources, one's argument is not necessarily more true than the
other, requiring comparison rather than verification. In this study, we propose
ClaimDiff, a novel dataset that primarily focuses on comparing the nuance
between claim pairs. In ClaimDiff, we provide 2,941 annotated claim pairs from
268 news articles. We observe that while humans are capable of detecting the
nuances between claims, strong baselines struggle to detect them, showing over
a 19% absolute gap with the humans. We hope this initial study could help
readers to gain an unbiased grasp of contentious issues through machine-aided
comparison.
| [
{
"version": "v1",
"created": "Tue, 24 May 2022 17:29:13 GMT"
},
{
"version": "v2",
"created": "Sun, 11 Jun 2023 05:30:31 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Ko",
"Miyoung",
""
],
[
"Seong",
"Ingyu",
""
],
[
"Lee",
"Hwaran",
""
],
[
"Park",
"Joonsuk",
""
],
[
"Chang",
"Minsuk",
""
],
[
"Seo",
"Minjoon",
""
]
] | TITLE: ClaimDiff: Comparing and Contrasting Claims on Contentious Issues
ABSTRACT: With the growing importance of detecting misinformation, many studies have
focused on verifying factual claims by retrieving evidence. However, canonical
fact verification tasks do not apply to catching subtle differences in
factually consistent claims, which might still bias the readers, especially on
contentious political or economic issues. Our underlying assumption is that
among the trusted sources, one's argument is not necessarily more true than the
other, requiring comparison rather than verification. In this study, we propose
ClaimDiff, a novel dataset that primarily focuses on comparing the nuance
between claim pairs. In ClaimDiff, we provide 2,941 annotated claim pairs from
268 news articles. We observe that while humans are capable of detecting the
nuances between claims, strong baselines struggle to detect them, showing over
a 19% absolute gap with the humans. We hope this initial study could help
readers to gain an unbiased grasp of contentious issues through machine-aided
comparison.
|
2206.00621 | Yan Zeng | Yan Zeng, Wangchunshu Zhou, Ao Luo, Ziming Cheng, Xinsong Zhang | Cross-View Language Modeling: Towards Unified Cross-Lingual Cross-Modal
Pre-training | ACL 2023 | null | null | null | cs.CL cs.AI cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we introduce Cross-View Language Modeling, a simple and
effective pre-training framework that unifies cross-lingual and cross-modal
pre-training with shared architectures and objectives. Our approach is
motivated by a key observation that cross-lingual and cross-modal pre-training
share the same goal of aligning two different views of the same object into a
common semantic space. To this end, the cross-view language modeling framework
considers both multi-modal data (i.e., image-caption pairs) and multi-lingual
data (i.e., parallel sentence pairs) as two different views of the same object,
and trains the model to align the two views by maximizing the mutual
information between them with conditional masked language modeling and
contrastive learning. We pre-train CCLM, a Cross-lingual Cross-modal Language
Model, with the cross-view language modeling framework. Empirical results on
IGLUE, a multi-lingual multi-modal benchmark, and two multi-lingual image-text
retrieval datasets show that while conceptually simpler, CCLM significantly
outperforms the prior state-of-the-art with an average absolute improvement of
over 10%. Moreover, CCLM is the first multi-lingual multi-modal pre-trained
model that surpasses the translate-test performance of representative English
vision-language models by zero-shot cross-lingual transfer.
| [
{
"version": "v1",
"created": "Wed, 1 Jun 2022 16:45:24 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 12:47:16 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zeng",
"Yan",
""
],
[
"Zhou",
"Wangchunshu",
""
],
[
"Luo",
"Ao",
""
],
[
"Cheng",
"Ziming",
""
],
[
"Zhang",
"Xinsong",
""
]
] | TITLE: Cross-View Language Modeling: Towards Unified Cross-Lingual Cross-Modal
Pre-training
ABSTRACT: In this paper, we introduce Cross-View Language Modeling, a simple and
effective pre-training framework that unifies cross-lingual and cross-modal
pre-training with shared architectures and objectives. Our approach is
motivated by a key observation that cross-lingual and cross-modal pre-training
share the same goal of aligning two different views of the same object into a
common semantic space. To this end, the cross-view language modeling framework
considers both multi-modal data (i.e., image-caption pairs) and multi-lingual
data (i.e., parallel sentence pairs) as two different views of the same object,
and trains the model to align the two views by maximizing the mutual
information between them with conditional masked language modeling and
contrastive learning. We pre-train CCLM, a Cross-lingual Cross-modal Language
Model, with the cross-view language modeling framework. Empirical results on
IGLUE, a multi-lingual multi-modal benchmark, and two multi-lingual image-text
retrieval datasets show that while conceptually simpler, CCLM significantly
outperforms the prior state-of-the-art with an average absolute improvement of
over 10%. Moreover, CCLM is the first multi-lingual multi-modal pre-trained
model that surpasses the translate-test performance of representative English
vision-language models by zero-shot cross-lingual transfer.
|
2206.04615 | Jaehoon Lee | Aarohi Srivastava, Abhinav Rastogi, Abhishek Rao, Abu Awal Md Shoeb,
Abubakar Abid, Adam Fisch, Adam R. Brown, Adam Santoro, Aditya Gupta, Adri\`a
Garriga-Alonso, Agnieszka Kluska, Aitor Lewkowycz, Akshat Agarwal, Alethea
Power, Alex Ray, Alex Warstadt, Alexander W. Kocurek, Ali Safaya, Ali Tazarv,
Alice Xiang, Alicia Parrish, Allen Nie, Aman Hussain, Amanda Askell, Amanda
Dsouza, Ambrose Slone, Ameet Rahane, Anantharaman S. Iyer, Anders Andreassen,
Andrea Madotto, Andrea Santilli, Andreas Stuhlm\"uller, Andrew Dai, Andrew
La, Andrew Lampinen, Andy Zou, Angela Jiang, Angelica Chen, Anh Vuong,
Animesh Gupta, Anna Gottardi, Antonio Norelli, Anu Venkatesh, Arash
Gholamidavoodi, Arfa Tabassum, Arul Menezes, Arun Kirubarajan, Asher
Mullokandov, Ashish Sabharwal, Austin Herrick, Avia Efrat, Aykut Erdem, Ayla
Karaka\c{s}, B. Ryan Roberts, Bao Sheng Loe, Barret Zoph, Bart{\l}omiej
Bojanowski, Batuhan \"Ozyurt, Behnam Hedayatnia, Behnam Neyshabur, Benjamin
Inden, Benno Stein, Berk Ekmekci, Bill Yuchen Lin, Blake Howald, Bryan
Orinion, Cameron Diao, Cameron Dour, Catherine Stinson, Cedrick Argueta,
C\'esar Ferri Ram\'irez, Chandan Singh, Charles Rathkopf, Chenlin Meng,
Chitta Baral, Chiyu Wu, Chris Callison-Burch, Chris Waites, Christian Voigt,
Christopher D. Manning, Christopher Potts, Cindy Ramirez, Clara E. Rivera,
Clemencia Siro, Colin Raffel, Courtney Ashcraft, Cristina Garbacea, Damien
Sileo, Dan Garrette, Dan Hendrycks, Dan Kilman, Dan Roth, Daniel Freeman,
Daniel Khashabi, Daniel Levy, Daniel Mosegu\'i Gonz\'alez, Danielle Perszyk,
Danny Hernandez, Danqi Chen, Daphne Ippolito, Dar Gilboa, David Dohan, David
Drakard, David Jurgens, Debajyoti Datta, Deep Ganguli, Denis Emelin, Denis
Kleyko, Deniz Yuret, Derek Chen, Derek Tam, Dieuwke Hupkes, Diganta Misra,
Dilyar Buzan, Dimitri Coelho Mollo, Diyi Yang, Dong-Ho Lee, Dylan Schrader,
Ekaterina Shutova, Ekin Dogus Cubuk, Elad Segal, Eleanor Hagerman, Elizabeth
Barnes, Elizabeth Donoway, Ellie Pavlick, Emanuele Rodola, Emma Lam, Eric
Chu, Eric Tang, Erkut Erdem, Ernie Chang, Ethan A. Chi, Ethan Dyer, Ethan
Jerzak, Ethan Kim, Eunice Engefu Manyasi, Evgenii Zheltonozhskii, Fanyue Xia,
Fatemeh Siar, Fernando Mart\'inez-Plumed, Francesca Happ\'e, Francois
Chollet, Frieda Rong, Gaurav Mishra, Genta Indra Winata, Gerard de Melo,
Germ\'an Kruszewski, Giambattista Parascandolo, Giorgio Mariani, Gloria Wang,
Gonzalo Jaimovitch-L\'opez, Gregor Betz, Guy Gur-Ari, Hana Galijasevic,
Hannah Kim, Hannah Rashkin, Hannaneh Hajishirzi, Harsh Mehta, Hayden Bogar,
Henry Shevlin, Hinrich Sch\"utze, Hiromu Yakura, Hongming Zhang, Hugh Mee
Wong, Ian Ng, Isaac Noble, Jaap Jumelet, Jack Geissinger, Jackson Kernion,
Jacob Hilton, Jaehoon Lee, Jaime Fern\'andez Fisac, James B. Simon, James
Koppel, James Zheng, James Zou, Jan Koco\'n, Jana Thompson, Janelle
Wingfield, Jared Kaplan, Jarema Radom, Jascha Sohl-Dickstein, Jason Phang,
Jason Wei, Jason Yosinski, Jekaterina Novikova, Jelle Bosscher, Jennifer
Marsh, Jeremy Kim, Jeroen Taal, Jesse Engel, Jesujoba Alabi, Jiacheng Xu,
Jiaming Song, Jillian Tang, Joan Waweru, John Burden, John Miller, John U.
Balis, Jonathan Batchelder, Jonathan Berant, J\"org Frohberg, Jos Rozen, Jose
Hernandez-Orallo, Joseph Boudeman, Joseph Guerr, Joseph Jones, Joshua B.
Tenenbaum, Joshua S. Rule, Joyce Chua, Kamil Kanclerz, Karen Livescu, Karl
Krauth, Karthik Gopalakrishnan, Katerina Ignatyeva, Katja Markert, Kaustubh
D. Dhole, Kevin Gimpel, Kevin Omondi, Kory Mathewson, Kristen Chiafullo,
Ksenia Shkaruta, Kumar Shridhar, Kyle McDonell, Kyle Richardson, Laria
Reynolds, Leo Gao, Li Zhang, Liam Dugan, Lianhui Qin, Lidia
Contreras-Ochando, Louis-Philippe Morency, Luca Moschella, Lucas Lam, Lucy
Noble, Ludwig Schmidt, Luheng He, Luis Oliveros Col\'on, Luke Metz, L\"utfi
Kerem \c{S}enel, Maarten Bosma, Maarten Sap, Maartje ter Hoeve, Maheen
Farooqi, Manaal Faruqui, Mantas Mazeika, Marco Baturan, Marco Marelli, Marco
Maru, Maria Jose Ram\'irez Quintana, Marie Tolkiehn, Mario Giulianelli,
Martha Lewis, Martin Potthast, Matthew L. Leavitt, Matthias Hagen, M\'aty\'as
Schubert, Medina Orduna Baitemirova, Melody Arnaud, Melvin McElrath, Michael
A. Yee, Michael Cohen, Michael Gu, Michael Ivanitskiy, Michael Starritt,
Michael Strube, Micha{\l} Sw\k{e}drowski, Michele Bevilacqua, Michihiro
Yasunaga, Mihir Kale, Mike Cain, Mimee Xu, Mirac Suzgun, Mitch Walker, Mo
Tiwari, Mohit Bansal, Moin Aminnaseri, Mor Geva, Mozhdeh Gheini, Mukund Varma
T, Nanyun Peng, Nathan A. Chi, Nayeon Lee, Neta Gur-Ari Krakover, Nicholas
Cameron, Nicholas Roberts, Nick Doiron, Nicole Martinez, Nikita Nangia,
Niklas Deckers, Niklas Muennighoff, Nitish Shirish Keskar, Niveditha S. Iyer,
Noah Constant, Noah Fiedel, Nuan Wen, Oliver Zhang, Omar Agha, Omar
Elbaghdadi, Omer Levy, Owain Evans, Pablo Antonio Moreno Casares, Parth
Doshi, Pascale Fung, Paul Pu Liang, Paul Vicol, Pegah Alipoormolabashi,
Peiyuan Liao, Percy Liang, Peter Chang, Peter Eckersley, Phu Mon Htut, Pinyu
Hwang, Piotr Mi{\l}kowski, Piyush Patil, Pouya Pezeshkpour, Priti Oli,
Qiaozhu Mei, Qing Lyu, Qinlang Chen, Rabin Banjade, Rachel Etta Rudolph,
Raefer Gabriel, Rahel Habacker, Ramon Risco, Rapha\"el Milli\`ere, Rhythm
Garg, Richard Barnes, Rif A. Saurous, Riku Arakawa, Robbe Raymaekers, Robert
Frank, Rohan Sikand, Roman Novak, Roman Sitelew, Ronan LeBras, Rosanne Liu,
Rowan Jacobs, Rui Zhang, Ruslan Salakhutdinov, Ryan Chi, Ryan Lee, Ryan
Stovall, Ryan Teehan, Rylan Yang, Sahib Singh, Saif M. Mohammad, Sajant
Anand, Sam Dillavou, Sam Shleifer, Sam Wiseman, Samuel Gruetter, Samuel R.
Bowman, Samuel S. Schoenholz, Sanghyun Han, Sanjeev Kwatra, Sarah A. Rous,
Sarik Ghazarian, Sayan Ghosh, Sean Casey, Sebastian Bischoff, Sebastian
Gehrmann, Sebastian Schuster, Sepideh Sadeghi, Shadi Hamdan, Sharon Zhou,
Shashank Srivastava, Sherry Shi, Shikhar Singh, Shima Asaadi, Shixiang Shane
Gu, Shubh Pachchigar, Shubham Toshniwal, Shyam Upadhyay, Shyamolima (Shammie)
Debnath, Siamak Shakeri, Simon Thormeyer, Simone Melzi, Siva Reddy, Sneha
Priscilla Makini, Soo-Hwan Lee, Spencer Torene, Sriharsha Hatwar, Stanislas
Dehaene, Stefan Divic, Stefano Ermon, Stella Biderman, Stephanie Lin, Stephen
Prasad, Steven T. Piantadosi, Stuart M. Shieber, Summer Misherghi, Svetlana
Kiritchenko, Swaroop Mishra, Tal Linzen, Tal Schuster, Tao Li, Tao Yu, Tariq
Ali, Tatsu Hashimoto, Te-Lin Wu, Th\'eo Desbordes, Theodore Rothschild,
Thomas Phan, Tianle Wang, Tiberius Nkinyili, Timo Schick, Timofei Kornev,
Titus Tunduny, Tobias Gerstenberg, Trenton Chang, Trishala Neeraj, Tushar
Khot, Tyler Shultz, Uri Shaham, Vedant Misra, Vera Demberg, Victoria Nyamai,
Vikas Raunak, Vinay Ramasesh, Vinay Uday Prabhu, Vishakh Padmakumar, Vivek
Srikumar, William Fedus, William Saunders, William Zhang, Wout Vossen, Xiang
Ren, Xiaoyu Tong, Xinran Zhao, Xinyi Wu, Xudong Shen, Yadollah Yaghoobzadeh,
Yair Lakretz, Yangqiu Song, Yasaman Bahri, Yejin Choi, Yichi Yang, Yiding
Hao, Yifu Chen, Yonatan Belinkov, Yu Hou, Yufang Hou, Yuntao Bai, Zachary
Seid, Zhuoye Zhao, Zijian Wang, Zijie J. Wang, Zirui Wang, Ziyi Wu | Beyond the Imitation Game: Quantifying and extrapolating the
capabilities of language models | 27 pages, 17 figures + references and appendices, repo:
https://github.com/google/BIG-bench | Transactions on Machine Learning Research, May/2022,
https://openreview.net/forum?id=uyTL5Bvosj | null | null | cs.CL cs.AI cs.CY cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Language models demonstrate both quantitative improvement and new qualitative
capabilities with increasing scale. Despite their potentially transformative
impact, these new capabilities are as yet poorly characterized. In order to
inform future research, prepare for disruptive new model capabilities, and
ameliorate socially harmful effects, it is vital that we understand the present
and near-future capabilities and limitations of language models. To address
this challenge, we introduce the Beyond the Imitation Game benchmark
(BIG-bench). BIG-bench currently consists of 204 tasks, contributed by 450
authors across 132 institutions. Task topics are diverse, drawing problems from
linguistics, childhood development, math, common-sense reasoning, biology,
physics, social bias, software development, and beyond. BIG-bench focuses on
tasks that are believed to be beyond the capabilities of current language
models. We evaluate the behavior of OpenAI's GPT models, Google-internal dense
transformer architectures, and Switch-style sparse transformers on BIG-bench,
across model sizes spanning millions to hundreds of billions of parameters. In
addition, a team of human expert raters performed all tasks in order to provide
a strong baseline. Findings include: model performance and calibration both
improve with scale, but are poor in absolute terms (and when compared with
rater performance); performance is remarkably similar across model classes,
though with benefits from sparsity; tasks that improve gradually and
predictably commonly involve a large knowledge or memorization component,
whereas tasks that exhibit "breakthrough" behavior at a critical scale often
involve multiple steps or components, or brittle metrics; social bias typically
increases with scale in settings with ambiguous context, but this can be
improved with prompting.
| [
{
"version": "v1",
"created": "Thu, 9 Jun 2022 17:05:34 GMT"
},
{
"version": "v2",
"created": "Fri, 10 Jun 2022 17:17:09 GMT"
},
{
"version": "v3",
"created": "Mon, 12 Jun 2023 17:51:15 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Srivastava",
"Aarohi",
"",
"Shammie"
],
[
"Rastogi",
"Abhinav",
"",
"Shammie"
],
[
"Rao",
"Abhishek",
"",
"Shammie"
],
[
"Shoeb",
"Abu Awal Md",
"",
"Shammie"
],
[
"Abid",
"Abubakar",
"",
"Shammie"
],
[
"Fisch",
"Adam",
"",
"Shammie"
],
[
"Brown",
"Adam R.",
"",
"Shammie"
],
[
"Santoro",
"Adam",
"",
"Shammie"
],
[
"Gupta",
"Aditya",
"",
"Shammie"
],
[
"Garriga-Alonso",
"Adrià",
"",
"Shammie"
],
[
"Kluska",
"Agnieszka",
"",
"Shammie"
],
[
"Lewkowycz",
"Aitor",
"",
"Shammie"
],
[
"Agarwal",
"Akshat",
"",
"Shammie"
],
[
"Power",
"Alethea",
"",
"Shammie"
],
[
"Ray",
"Alex",
"",
"Shammie"
],
[
"Warstadt",
"Alex",
"",
"Shammie"
],
[
"Kocurek",
"Alexander W.",
"",
"Shammie"
],
[
"Safaya",
"Ali",
"",
"Shammie"
],
[
"Tazarv",
"Ali",
"",
"Shammie"
],
[
"Xiang",
"Alice",
"",
"Shammie"
],
[
"Parrish",
"Alicia",
"",
"Shammie"
],
[
"Nie",
"Allen",
"",
"Shammie"
],
[
"Hussain",
"Aman",
"",
"Shammie"
],
[
"Askell",
"Amanda",
"",
"Shammie"
],
[
"Dsouza",
"Amanda",
"",
"Shammie"
],
[
"Slone",
"Ambrose",
"",
"Shammie"
],
[
"Rahane",
"Ameet",
"",
"Shammie"
],
[
"Iyer",
"Anantharaman S.",
"",
"Shammie"
],
[
"Andreassen",
"Anders",
"",
"Shammie"
],
[
"Madotto",
"Andrea",
"",
"Shammie"
],
[
"Santilli",
"Andrea",
"",
"Shammie"
],
[
"Stuhlmüller",
"Andreas",
"",
"Shammie"
],
[
"Dai",
"Andrew",
"",
"Shammie"
],
[
"La",
"Andrew",
"",
"Shammie"
],
[
"Lampinen",
"Andrew",
"",
"Shammie"
],
[
"Zou",
"Andy",
"",
"Shammie"
],
[
"Jiang",
"Angela",
"",
"Shammie"
],
[
"Chen",
"Angelica",
"",
"Shammie"
],
[
"Vuong",
"Anh",
"",
"Shammie"
],
[
"Gupta",
"Animesh",
"",
"Shammie"
],
[
"Gottardi",
"Anna",
"",
"Shammie"
],
[
"Norelli",
"Antonio",
"",
"Shammie"
],
[
"Venkatesh",
"Anu",
"",
"Shammie"
],
[
"Gholamidavoodi",
"Arash",
"",
"Shammie"
],
[
"Tabassum",
"Arfa",
"",
"Shammie"
],
[
"Menezes",
"Arul",
"",
"Shammie"
],
[
"Kirubarajan",
"Arun",
"",
"Shammie"
],
[
"Mullokandov",
"Asher",
"",
"Shammie"
],
[
"Sabharwal",
"Ashish",
"",
"Shammie"
],
[
"Herrick",
"Austin",
"",
"Shammie"
],
[
"Efrat",
"Avia",
"",
"Shammie"
],
[
"Erdem",
"Aykut",
"",
"Shammie"
],
[
"Karakaş",
"Ayla",
"",
"Shammie"
],
[
"Roberts",
"B. Ryan",
"",
"Shammie"
],
[
"Loe",
"Bao Sheng",
"",
"Shammie"
],
[
"Zoph",
"Barret",
"",
"Shammie"
],
[
"Bojanowski",
"Bartłomiej",
"",
"Shammie"
],
[
"Özyurt",
"Batuhan",
"",
"Shammie"
],
[
"Hedayatnia",
"Behnam",
"",
"Shammie"
],
[
"Neyshabur",
"Behnam",
"",
"Shammie"
],
[
"Inden",
"Benjamin",
"",
"Shammie"
],
[
"Stein",
"Benno",
"",
"Shammie"
],
[
"Ekmekci",
"Berk",
"",
"Shammie"
],
[
"Lin",
"Bill Yuchen",
"",
"Shammie"
],
[
"Howald",
"Blake",
"",
"Shammie"
],
[
"Orinion",
"Bryan",
"",
"Shammie"
],
[
"Diao",
"Cameron",
"",
"Shammie"
],
[
"Dour",
"Cameron",
"",
"Shammie"
],
[
"Stinson",
"Catherine",
"",
"Shammie"
],
[
"Argueta",
"Cedrick",
"",
"Shammie"
],
[
"Ramírez",
"César Ferri",
"",
"Shammie"
],
[
"Singh",
"Chandan",
"",
"Shammie"
],
[
"Rathkopf",
"Charles",
"",
"Shammie"
],
[
"Meng",
"Chenlin",
"",
"Shammie"
],
[
"Baral",
"Chitta",
"",
"Shammie"
],
[
"Wu",
"Chiyu",
"",
"Shammie"
],
[
"Callison-Burch",
"Chris",
"",
"Shammie"
],
[
"Waites",
"Chris",
"",
"Shammie"
],
[
"Voigt",
"Christian",
"",
"Shammie"
],
[
"Manning",
"Christopher D.",
"",
"Shammie"
],
[
"Potts",
"Christopher",
"",
"Shammie"
],
[
"Ramirez",
"Cindy",
"",
"Shammie"
],
[
"Rivera",
"Clara E.",
"",
"Shammie"
],
[
"Siro",
"Clemencia",
"",
"Shammie"
],
[
"Raffel",
"Colin",
"",
"Shammie"
],
[
"Ashcraft",
"Courtney",
"",
"Shammie"
],
[
"Garbacea",
"Cristina",
"",
"Shammie"
],
[
"Sileo",
"Damien",
"",
"Shammie"
],
[
"Garrette",
"Dan",
"",
"Shammie"
],
[
"Hendrycks",
"Dan",
"",
"Shammie"
],
[
"Kilman",
"Dan",
"",
"Shammie"
],
[
"Roth",
"Dan",
"",
"Shammie"
],
[
"Freeman",
"Daniel",
"",
"Shammie"
],
[
"Khashabi",
"Daniel",
"",
"Shammie"
],
[
"Levy",
"Daniel",
"",
"Shammie"
],
[
"González",
"Daniel Moseguí",
"",
"Shammie"
],
[
"Perszyk",
"Danielle",
"",
"Shammie"
],
[
"Hernandez",
"Danny",
"",
"Shammie"
],
[
"Chen",
"Danqi",
"",
"Shammie"
],
[
"Ippolito",
"Daphne",
"",
"Shammie"
],
[
"Gilboa",
"Dar",
"",
"Shammie"
],
[
"Dohan",
"David",
"",
"Shammie"
],
[
"Drakard",
"David",
"",
"Shammie"
],
[
"Jurgens",
"David",
"",
"Shammie"
],
[
"Datta",
"Debajyoti",
"",
"Shammie"
],
[
"Ganguli",
"Deep",
"",
"Shammie"
],
[
"Emelin",
"Denis",
"",
"Shammie"
],
[
"Kleyko",
"Denis",
"",
"Shammie"
],
[
"Yuret",
"Deniz",
"",
"Shammie"
],
[
"Chen",
"Derek",
"",
"Shammie"
],
[
"Tam",
"Derek",
"",
"Shammie"
],
[
"Hupkes",
"Dieuwke",
"",
"Shammie"
],
[
"Misra",
"Diganta",
"",
"Shammie"
],
[
"Buzan",
"Dilyar",
"",
"Shammie"
],
[
"Mollo",
"Dimitri Coelho",
"",
"Shammie"
],
[
"Yang",
"Diyi",
"",
"Shammie"
],
[
"Lee",
"Dong-Ho",
"",
"Shammie"
],
[
"Schrader",
"Dylan",
"",
"Shammie"
],
[
"Shutova",
"Ekaterina",
"",
"Shammie"
],
[
"Cubuk",
"Ekin Dogus",
"",
"Shammie"
],
[
"Segal",
"Elad",
"",
"Shammie"
],
[
"Hagerman",
"Eleanor",
"",
"Shammie"
],
[
"Barnes",
"Elizabeth",
"",
"Shammie"
],
[
"Donoway",
"Elizabeth",
"",
"Shammie"
],
[
"Pavlick",
"Ellie",
"",
"Shammie"
],
[
"Rodola",
"Emanuele",
"",
"Shammie"
],
[
"Lam",
"Emma",
"",
"Shammie"
],
[
"Chu",
"Eric",
"",
"Shammie"
],
[
"Tang",
"Eric",
"",
"Shammie"
],
[
"Erdem",
"Erkut",
"",
"Shammie"
],
[
"Chang",
"Ernie",
"",
"Shammie"
],
[
"Chi",
"Ethan A.",
"",
"Shammie"
],
[
"Dyer",
"Ethan",
"",
"Shammie"
],
[
"Jerzak",
"Ethan",
"",
"Shammie"
],
[
"Kim",
"Ethan",
"",
"Shammie"
],
[
"Manyasi",
"Eunice Engefu",
"",
"Shammie"
],
[
"Zheltonozhskii",
"Evgenii",
"",
"Shammie"
],
[
"Xia",
"Fanyue",
"",
"Shammie"
],
[
"Siar",
"Fatemeh",
"",
"Shammie"
],
[
"Martínez-Plumed",
"Fernando",
"",
"Shammie"
],
[
"Happé",
"Francesca",
"",
"Shammie"
],
[
"Chollet",
"Francois",
"",
"Shammie"
],
[
"Rong",
"Frieda",
"",
"Shammie"
],
[
"Mishra",
"Gaurav",
"",
"Shammie"
],
[
"Winata",
"Genta Indra",
"",
"Shammie"
],
[
"de Melo",
"Gerard",
"",
"Shammie"
],
[
"Kruszewski",
"Germán",
"",
"Shammie"
],
[
"Parascandolo",
"Giambattista",
"",
"Shammie"
],
[
"Mariani",
"Giorgio",
"",
"Shammie"
],
[
"Wang",
"Gloria",
"",
"Shammie"
],
[
"Jaimovitch-López",
"Gonzalo",
"",
"Shammie"
],
[
"Betz",
"Gregor",
"",
"Shammie"
],
[
"Gur-Ari",
"Guy",
"",
"Shammie"
],
[
"Galijasevic",
"Hana",
"",
"Shammie"
],
[
"Kim",
"Hannah",
"",
"Shammie"
],
[
"Rashkin",
"Hannah",
"",
"Shammie"
],
[
"Hajishirzi",
"Hannaneh",
"",
"Shammie"
],
[
"Mehta",
"Harsh",
"",
"Shammie"
],
[
"Bogar",
"Hayden",
"",
"Shammie"
],
[
"Shevlin",
"Henry",
"",
"Shammie"
],
[
"Schütze",
"Hinrich",
"",
"Shammie"
],
[
"Yakura",
"Hiromu",
"",
"Shammie"
],
[
"Zhang",
"Hongming",
"",
"Shammie"
],
[
"Wong",
"Hugh Mee",
"",
"Shammie"
],
[
"Ng",
"Ian",
"",
"Shammie"
],
[
"Noble",
"Isaac",
"",
"Shammie"
],
[
"Jumelet",
"Jaap",
"",
"Shammie"
],
[
"Geissinger",
"Jack",
"",
"Shammie"
],
[
"Kernion",
"Jackson",
"",
"Shammie"
],
[
"Hilton",
"Jacob",
"",
"Shammie"
],
[
"Lee",
"Jaehoon",
"",
"Shammie"
],
[
"Fisac",
"Jaime Fernández",
"",
"Shammie"
],
[
"Simon",
"James B.",
"",
"Shammie"
],
[
"Koppel",
"James",
"",
"Shammie"
],
[
"Zheng",
"James",
"",
"Shammie"
],
[
"Zou",
"James",
"",
"Shammie"
],
[
"Kocoń",
"Jan",
"",
"Shammie"
],
[
"Thompson",
"Jana",
"",
"Shammie"
],
[
"Wingfield",
"Janelle",
"",
"Shammie"
],
[
"Kaplan",
"Jared",
"",
"Shammie"
],
[
"Radom",
"Jarema",
"",
"Shammie"
],
[
"Sohl-Dickstein",
"Jascha",
"",
"Shammie"
],
[
"Phang",
"Jason",
"",
"Shammie"
],
[
"Wei",
"Jason",
"",
"Shammie"
],
[
"Yosinski",
"Jason",
"",
"Shammie"
],
[
"Novikova",
"Jekaterina",
"",
"Shammie"
],
[
"Bosscher",
"Jelle",
"",
"Shammie"
],
[
"Marsh",
"Jennifer",
"",
"Shammie"
],
[
"Kim",
"Jeremy",
"",
"Shammie"
],
[
"Taal",
"Jeroen",
"",
"Shammie"
],
[
"Engel",
"Jesse",
"",
"Shammie"
],
[
"Alabi",
"Jesujoba",
"",
"Shammie"
],
[
"Xu",
"Jiacheng",
"",
"Shammie"
],
[
"Song",
"Jiaming",
"",
"Shammie"
],
[
"Tang",
"Jillian",
"",
"Shammie"
],
[
"Waweru",
"Joan",
"",
"Shammie"
],
[
"Burden",
"John",
"",
"Shammie"
],
[
"Miller",
"John",
"",
"Shammie"
],
[
"Balis",
"John U.",
"",
"Shammie"
],
[
"Batchelder",
"Jonathan",
"",
"Shammie"
],
[
"Berant",
"Jonathan",
"",
"Shammie"
],
[
"Frohberg",
"Jörg",
"",
"Shammie"
],
[
"Rozen",
"Jos",
"",
"Shammie"
],
[
"Hernandez-Orallo",
"Jose",
"",
"Shammie"
],
[
"Boudeman",
"Joseph",
"",
"Shammie"
],
[
"Guerr",
"Joseph",
"",
"Shammie"
],
[
"Jones",
"Joseph",
"",
"Shammie"
],
[
"Tenenbaum",
"Joshua B.",
"",
"Shammie"
],
[
"Rule",
"Joshua S.",
"",
"Shammie"
],
[
"Chua",
"Joyce",
"",
"Shammie"
],
[
"Kanclerz",
"Kamil",
"",
"Shammie"
],
[
"Livescu",
"Karen",
"",
"Shammie"
],
[
"Krauth",
"Karl",
"",
"Shammie"
],
[
"Gopalakrishnan",
"Karthik",
"",
"Shammie"
],
[
"Ignatyeva",
"Katerina",
"",
"Shammie"
],
[
"Markert",
"Katja",
"",
"Shammie"
],
[
"Dhole",
"Kaustubh D.",
"",
"Shammie"
],
[
"Gimpel",
"Kevin",
"",
"Shammie"
],
[
"Omondi",
"Kevin",
"",
"Shammie"
],
[
"Mathewson",
"Kory",
"",
"Shammie"
],
[
"Chiafullo",
"Kristen",
"",
"Shammie"
],
[
"Shkaruta",
"Ksenia",
"",
"Shammie"
],
[
"Shridhar",
"Kumar",
"",
"Shammie"
],
[
"McDonell",
"Kyle",
"",
"Shammie"
],
[
"Richardson",
"Kyle",
"",
"Shammie"
],
[
"Reynolds",
"Laria",
"",
"Shammie"
],
[
"Gao",
"Leo",
"",
"Shammie"
],
[
"Zhang",
"Li",
"",
"Shammie"
],
[
"Dugan",
"Liam",
"",
"Shammie"
],
[
"Qin",
"Lianhui",
"",
"Shammie"
],
[
"Contreras-Ochando",
"Lidia",
"",
"Shammie"
],
[
"Morency",
"Louis-Philippe",
"",
"Shammie"
],
[
"Moschella",
"Luca",
"",
"Shammie"
],
[
"Lam",
"Lucas",
"",
"Shammie"
],
[
"Noble",
"Lucy",
"",
"Shammie"
],
[
"Schmidt",
"Ludwig",
"",
"Shammie"
],
[
"He",
"Luheng",
"",
"Shammie"
],
[
"Colón",
"Luis Oliveros",
"",
"Shammie"
],
[
"Metz",
"Luke",
"",
"Shammie"
],
[
"Şenel",
"Lütfi Kerem",
"",
"Shammie"
],
[
"Bosma",
"Maarten",
"",
"Shammie"
],
[
"Sap",
"Maarten",
"",
"Shammie"
],
[
"ter Hoeve",
"Maartje",
"",
"Shammie"
],
[
"Farooqi",
"Maheen",
"",
"Shammie"
],
[
"Faruqui",
"Manaal",
"",
"Shammie"
],
[
"Mazeika",
"Mantas",
"",
"Shammie"
],
[
"Baturan",
"Marco",
"",
"Shammie"
],
[
"Marelli",
"Marco",
"",
"Shammie"
],
[
"Maru",
"Marco",
"",
"Shammie"
],
[
"Quintana",
"Maria Jose Ramírez",
"",
"Shammie"
],
[
"Tolkiehn",
"Marie",
"",
"Shammie"
],
[
"Giulianelli",
"Mario",
"",
"Shammie"
],
[
"Lewis",
"Martha",
"",
"Shammie"
],
[
"Potthast",
"Martin",
"",
"Shammie"
],
[
"Leavitt",
"Matthew L.",
"",
"Shammie"
],
[
"Hagen",
"Matthias",
"",
"Shammie"
],
[
"Schubert",
"Mátyás",
"",
"Shammie"
],
[
"Baitemirova",
"Medina Orduna",
"",
"Shammie"
],
[
"Arnaud",
"Melody",
"",
"Shammie"
],
[
"McElrath",
"Melvin",
"",
"Shammie"
],
[
"Yee",
"Michael A.",
"",
"Shammie"
],
[
"Cohen",
"Michael",
"",
"Shammie"
],
[
"Gu",
"Michael",
"",
"Shammie"
],
[
"Ivanitskiy",
"Michael",
"",
"Shammie"
],
[
"Starritt",
"Michael",
"",
"Shammie"
],
[
"Strube",
"Michael",
"",
"Shammie"
],
[
"Swędrowski",
"Michał",
"",
"Shammie"
],
[
"Bevilacqua",
"Michele",
"",
"Shammie"
],
[
"Yasunaga",
"Michihiro",
"",
"Shammie"
],
[
"Kale",
"Mihir",
"",
"Shammie"
],
[
"Cain",
"Mike",
"",
"Shammie"
],
[
"Xu",
"Mimee",
"",
"Shammie"
],
[
"Suzgun",
"Mirac",
"",
"Shammie"
],
[
"Walker",
"Mitch",
"",
"Shammie"
],
[
"Tiwari",
"Mo",
"",
"Shammie"
],
[
"Bansal",
"Mohit",
"",
"Shammie"
],
[
"Aminnaseri",
"Moin",
"",
"Shammie"
],
[
"Geva",
"Mor",
"",
"Shammie"
],
[
"Gheini",
"Mozhdeh",
"",
"Shammie"
],
[
"T",
"Mukund Varma",
"",
"Shammie"
],
[
"Peng",
"Nanyun",
"",
"Shammie"
],
[
"Chi",
"Nathan A.",
"",
"Shammie"
],
[
"Lee",
"Nayeon",
"",
"Shammie"
],
[
"Krakover",
"Neta Gur-Ari",
"",
"Shammie"
],
[
"Cameron",
"Nicholas",
"",
"Shammie"
],
[
"Roberts",
"Nicholas",
"",
"Shammie"
],
[
"Doiron",
"Nick",
"",
"Shammie"
],
[
"Martinez",
"Nicole",
"",
"Shammie"
],
[
"Nangia",
"Nikita",
"",
"Shammie"
],
[
"Deckers",
"Niklas",
"",
"Shammie"
],
[
"Muennighoff",
"Niklas",
"",
"Shammie"
],
[
"Keskar",
"Nitish Shirish",
"",
"Shammie"
],
[
"Iyer",
"Niveditha S.",
"",
"Shammie"
],
[
"Constant",
"Noah",
"",
"Shammie"
],
[
"Fiedel",
"Noah",
"",
"Shammie"
],
[
"Wen",
"Nuan",
"",
"Shammie"
],
[
"Zhang",
"Oliver",
"",
"Shammie"
],
[
"Agha",
"Omar",
"",
"Shammie"
],
[
"Elbaghdadi",
"Omar",
"",
"Shammie"
],
[
"Levy",
"Omer",
"",
"Shammie"
],
[
"Evans",
"Owain",
"",
"Shammie"
],
[
"Casares",
"Pablo Antonio Moreno",
"",
"Shammie"
],
[
"Doshi",
"Parth",
"",
"Shammie"
],
[
"Fung",
"Pascale",
"",
"Shammie"
],
[
"Liang",
"Paul Pu",
"",
"Shammie"
],
[
"Vicol",
"Paul",
"",
"Shammie"
],
[
"Alipoormolabashi",
"Pegah",
"",
"Shammie"
],
[
"Liao",
"Peiyuan",
"",
"Shammie"
],
[
"Liang",
"Percy",
"",
"Shammie"
],
[
"Chang",
"Peter",
"",
"Shammie"
],
[
"Eckersley",
"Peter",
"",
"Shammie"
],
[
"Htut",
"Phu Mon",
"",
"Shammie"
],
[
"Hwang",
"Pinyu",
"",
"Shammie"
],
[
"Miłkowski",
"Piotr",
"",
"Shammie"
],
[
"Patil",
"Piyush",
"",
"Shammie"
],
[
"Pezeshkpour",
"Pouya",
"",
"Shammie"
],
[
"Oli",
"Priti",
"",
"Shammie"
],
[
"Mei",
"Qiaozhu",
"",
"Shammie"
],
[
"Lyu",
"Qing",
"",
"Shammie"
],
[
"Chen",
"Qinlang",
"",
"Shammie"
],
[
"Banjade",
"Rabin",
"",
"Shammie"
],
[
"Rudolph",
"Rachel Etta",
"",
"Shammie"
],
[
"Gabriel",
"Raefer",
"",
"Shammie"
],
[
"Habacker",
"Rahel",
"",
"Shammie"
],
[
"Risco",
"Ramon",
"",
"Shammie"
],
[
"Millière",
"Raphaël",
"",
"Shammie"
],
[
"Garg",
"Rhythm",
"",
"Shammie"
],
[
"Barnes",
"Richard",
"",
"Shammie"
],
[
"Saurous",
"Rif A.",
"",
"Shammie"
],
[
"Arakawa",
"Riku",
"",
"Shammie"
],
[
"Raymaekers",
"Robbe",
"",
"Shammie"
],
[
"Frank",
"Robert",
"",
"Shammie"
],
[
"Sikand",
"Rohan",
"",
"Shammie"
],
[
"Novak",
"Roman",
"",
"Shammie"
],
[
"Sitelew",
"Roman",
"",
"Shammie"
],
[
"LeBras",
"Ronan",
"",
"Shammie"
],
[
"Liu",
"Rosanne",
"",
"Shammie"
],
[
"Jacobs",
"Rowan",
"",
"Shammie"
],
[
"Zhang",
"Rui",
"",
"Shammie"
],
[
"Salakhutdinov",
"Ruslan",
"",
"Shammie"
],
[
"Chi",
"Ryan",
"",
"Shammie"
],
[
"Lee",
"Ryan",
"",
"Shammie"
],
[
"Stovall",
"Ryan",
"",
"Shammie"
],
[
"Teehan",
"Ryan",
"",
"Shammie"
],
[
"Yang",
"Rylan",
"",
"Shammie"
],
[
"Singh",
"Sahib",
"",
"Shammie"
],
[
"Mohammad",
"Saif M.",
"",
"Shammie"
],
[
"Anand",
"Sajant",
"",
"Shammie"
],
[
"Dillavou",
"Sam",
"",
"Shammie"
],
[
"Shleifer",
"Sam",
"",
"Shammie"
],
[
"Wiseman",
"Sam",
"",
"Shammie"
],
[
"Gruetter",
"Samuel",
"",
"Shammie"
],
[
"Bowman",
"Samuel R.",
"",
"Shammie"
],
[
"Schoenholz",
"Samuel S.",
"",
"Shammie"
],
[
"Han",
"Sanghyun",
"",
"Shammie"
],
[
"Kwatra",
"Sanjeev",
"",
"Shammie"
],
[
"Rous",
"Sarah A.",
"",
"Shammie"
],
[
"Ghazarian",
"Sarik",
"",
"Shammie"
],
[
"Ghosh",
"Sayan",
"",
"Shammie"
],
[
"Casey",
"Sean",
"",
"Shammie"
],
[
"Bischoff",
"Sebastian",
"",
"Shammie"
],
[
"Gehrmann",
"Sebastian",
"",
"Shammie"
],
[
"Schuster",
"Sebastian",
"",
"Shammie"
],
[
"Sadeghi",
"Sepideh",
"",
"Shammie"
],
[
"Hamdan",
"Shadi",
"",
"Shammie"
],
[
"Zhou",
"Sharon",
"",
"Shammie"
],
[
"Srivastava",
"Shashank",
"",
"Shammie"
],
[
"Shi",
"Sherry",
"",
"Shammie"
],
[
"Singh",
"Shikhar",
"",
"Shammie"
],
[
"Asaadi",
"Shima",
"",
"Shammie"
],
[
"Gu",
"Shixiang Shane",
"",
"Shammie"
],
[
"Pachchigar",
"Shubh",
"",
"Shammie"
],
[
"Toshniwal",
"Shubham",
"",
"Shammie"
],
[
"Upadhyay",
"Shyam",
"",
"Shammie"
],
[
"Shyamolima",
"",
"",
"Shammie"
],
[
"Debnath",
"",
""
],
[
"Shakeri",
"Siamak",
""
],
[
"Thormeyer",
"Simon",
""
],
[
"Melzi",
"Simone",
""
],
[
"Reddy",
"Siva",
""
],
[
"Makini",
"Sneha Priscilla",
""
],
[
"Lee",
"Soo-Hwan",
""
],
[
"Torene",
"Spencer",
""
],
[
"Hatwar",
"Sriharsha",
""
],
[
"Dehaene",
"Stanislas",
""
],
[
"Divic",
"Stefan",
""
],
[
"Ermon",
"Stefano",
""
],
[
"Biderman",
"Stella",
""
],
[
"Lin",
"Stephanie",
""
],
[
"Prasad",
"Stephen",
""
],
[
"Piantadosi",
"Steven T.",
""
],
[
"Shieber",
"Stuart M.",
""
],
[
"Misherghi",
"Summer",
""
],
[
"Kiritchenko",
"Svetlana",
""
],
[
"Mishra",
"Swaroop",
""
],
[
"Linzen",
"Tal",
""
],
[
"Schuster",
"Tal",
""
],
[
"Li",
"Tao",
""
],
[
"Yu",
"Tao",
""
],
[
"Ali",
"Tariq",
""
],
[
"Hashimoto",
"Tatsu",
""
],
[
"Wu",
"Te-Lin",
""
],
[
"Desbordes",
"Théo",
""
],
[
"Rothschild",
"Theodore",
""
],
[
"Phan",
"Thomas",
""
],
[
"Wang",
"Tianle",
""
],
[
"Nkinyili",
"Tiberius",
""
],
[
"Schick",
"Timo",
""
],
[
"Kornev",
"Timofei",
""
],
[
"Tunduny",
"Titus",
""
],
[
"Gerstenberg",
"Tobias",
""
],
[
"Chang",
"Trenton",
""
],
[
"Neeraj",
"Trishala",
""
],
[
"Khot",
"Tushar",
""
],
[
"Shultz",
"Tyler",
""
],
[
"Shaham",
"Uri",
""
],
[
"Misra",
"Vedant",
""
],
[
"Demberg",
"Vera",
""
],
[
"Nyamai",
"Victoria",
""
],
[
"Raunak",
"Vikas",
""
],
[
"Ramasesh",
"Vinay",
""
],
[
"Prabhu",
"Vinay Uday",
""
],
[
"Padmakumar",
"Vishakh",
""
],
[
"Srikumar",
"Vivek",
""
],
[
"Fedus",
"William",
""
],
[
"Saunders",
"William",
""
],
[
"Zhang",
"William",
""
],
[
"Vossen",
"Wout",
""
],
[
"Ren",
"Xiang",
""
],
[
"Tong",
"Xiaoyu",
""
],
[
"Zhao",
"Xinran",
""
],
[
"Wu",
"Xinyi",
""
],
[
"Shen",
"Xudong",
""
],
[
"Yaghoobzadeh",
"Yadollah",
""
],
[
"Lakretz",
"Yair",
""
],
[
"Song",
"Yangqiu",
""
],
[
"Bahri",
"Yasaman",
""
],
[
"Choi",
"Yejin",
""
],
[
"Yang",
"Yichi",
""
],
[
"Hao",
"Yiding",
""
],
[
"Chen",
"Yifu",
""
],
[
"Belinkov",
"Yonatan",
""
],
[
"Hou",
"Yu",
""
],
[
"Hou",
"Yufang",
""
],
[
"Bai",
"Yuntao",
""
],
[
"Seid",
"Zachary",
""
],
[
"Zhao",
"Zhuoye",
""
],
[
"Wang",
"Zijian",
""
],
[
"Wang",
"Zijie J.",
""
],
[
"Wang",
"Zirui",
""
],
[
"Wu",
"Ziyi",
""
]
] | TITLE: Beyond the Imitation Game: Quantifying and extrapolating the
capabilities of language models
ABSTRACT: Language models demonstrate both quantitative improvement and new qualitative
capabilities with increasing scale. Despite their potentially transformative
impact, these new capabilities are as yet poorly characterized. In order to
inform future research, prepare for disruptive new model capabilities, and
ameliorate socially harmful effects, it is vital that we understand the present
and near-future capabilities and limitations of language models. To address
this challenge, we introduce the Beyond the Imitation Game benchmark
(BIG-bench). BIG-bench currently consists of 204 tasks, contributed by 450
authors across 132 institutions. Task topics are diverse, drawing problems from
linguistics, childhood development, math, common-sense reasoning, biology,
physics, social bias, software development, and beyond. BIG-bench focuses on
tasks that are believed to be beyond the capabilities of current language
models. We evaluate the behavior of OpenAI's GPT models, Google-internal dense
transformer architectures, and Switch-style sparse transformers on BIG-bench,
across model sizes spanning millions to hundreds of billions of parameters. In
addition, a team of human expert raters performed all tasks in order to provide
a strong baseline. Findings include: model performance and calibration both
improve with scale, but are poor in absolute terms (and when compared with
rater performance); performance is remarkably similar across model classes,
though with benefits from sparsity; tasks that improve gradually and
predictably commonly involve a large knowledge or memorization component,
whereas tasks that exhibit "breakthrough" behavior at a critical scale often
involve multiple steps or components, or brittle metrics; social bias typically
increases with scale in settings with ambiguous context, but this can be
improved with prompting.
|
2206.06354 | Jeroen Berrevoets | Jeroen Berrevoets, Nabeel Seedat, Fergus Imrie, Mihaela van der Schaar | Differentiable and Transportable Structure Learning | Accepted at the International Conference on Machine Learning (ICML)
2023 | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Directed acyclic graphs (DAGs) encode a lot of information about a particular
distribution in their structure. However, compute required to infer these
structures is typically super-exponential in the number of variables, as
inference requires a sweep of a combinatorially large space of potential
structures. That is, until recent advances made it possible to search this
space using a differentiable metric, drastically reducing search time. While
this technique -- named NOTEARS -- is widely considered a seminal work in
DAG-discovery, it concedes an important property in favour of
differentiability: transportability. To be transportable, the structures
discovered on one dataset must apply to another dataset from the same domain.
We introduce D-Struct which recovers transportability in the discovered
structures through a novel architecture and loss function while remaining fully
differentiable. Because D-Struct remains differentiable, our method can be
easily adopted in existing differentiable architectures, as was previously done
with NOTEARS. In our experiments, we empirically validate D-Struct with respect
to edge accuracy and structural Hamming distance in a variety of settings.
| [
{
"version": "v1",
"created": "Mon, 13 Jun 2022 17:50:53 GMT"
},
{
"version": "v2",
"created": "Thu, 29 Sep 2022 14:38:00 GMT"
},
{
"version": "v3",
"created": "Fri, 2 Jun 2023 09:27:10 GMT"
},
{
"version": "v4",
"created": "Mon, 12 Jun 2023 10:22:27 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Berrevoets",
"Jeroen",
""
],
[
"Seedat",
"Nabeel",
""
],
[
"Imrie",
"Fergus",
""
],
[
"van der Schaar",
"Mihaela",
""
]
] | TITLE: Differentiable and Transportable Structure Learning
ABSTRACT: Directed acyclic graphs (DAGs) encode a lot of information about a particular
distribution in their structure. However, compute required to infer these
structures is typically super-exponential in the number of variables, as
inference requires a sweep of a combinatorially large space of potential
structures. That is, until recent advances made it possible to search this
space using a differentiable metric, drastically reducing search time. While
this technique -- named NOTEARS -- is widely considered a seminal work in
DAG-discovery, it concedes an important property in favour of
differentiability: transportability. To be transportable, the structures
discovered on one dataset must apply to another dataset from the same domain.
We introduce D-Struct which recovers transportability in the discovered
structures through a novel architecture and loss function while remaining fully
differentiable. Because D-Struct remains differentiable, our method can be
easily adopted in existing differentiable architectures, as was previously done
with NOTEARS. In our experiments, we empirically validate D-Struct with respect
to edge accuracy and structural Hamming distance in a variety of settings.
|
2207.02621 | Jiahui Zhang | Jiahui Zhang and Fangneng Zhan and Rongliang Wu and Yingchen Yu and
Wenqing Zhang and Bai Song and Xiaoqin Zhang and Shijian Lu | VMRF: View Matching Neural Radiance Fields | This paper has been accepted to ACM MM 2022 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Neural Radiance Fields (NeRF) have demonstrated very impressive performance
in novel view synthesis via implicitly modelling 3D representations from
multi-view 2D images. However, most existing studies train NeRF models with
either reasonable camera pose initialization or manually-crafted camera pose
distributions which are often unavailable or hard to acquire in various
real-world data. We design VMRF, an innovative view matching NeRF that enables
effective NeRF training without requiring prior knowledge in camera poses or
camera pose distributions. VMRF introduces a view matching scheme, which
exploits unbalanced optimal transport to produce a feature transport plan for
mapping a rendered image with randomly initialized camera pose to the
corresponding real image. With the feature transport plan as the guidance, a
novel pose calibration technique is designed which rectifies the initially
randomized camera poses by predicting relative pose transformations between the
pair of rendered and real images. Extensive experiments over a number of
synthetic and real datasets show that the proposed VMRF outperforms the
state-of-the-art qualitatively and quantitatively by large margins.
| [
{
"version": "v1",
"created": "Wed, 6 Jul 2022 12:26:40 GMT"
},
{
"version": "v2",
"created": "Sun, 11 Jun 2023 18:40:36 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zhang",
"Jiahui",
""
],
[
"Zhan",
"Fangneng",
""
],
[
"Wu",
"Rongliang",
""
],
[
"Yu",
"Yingchen",
""
],
[
"Zhang",
"Wenqing",
""
],
[
"Song",
"Bai",
""
],
[
"Zhang",
"Xiaoqin",
""
],
[
"Lu",
"Shijian",
""
]
] | TITLE: VMRF: View Matching Neural Radiance Fields
ABSTRACT: Neural Radiance Fields (NeRF) have demonstrated very impressive performance
in novel view synthesis via implicitly modelling 3D representations from
multi-view 2D images. However, most existing studies train NeRF models with
either reasonable camera pose initialization or manually-crafted camera pose
distributions which are often unavailable or hard to acquire in various
real-world data. We design VMRF, an innovative view matching NeRF that enables
effective NeRF training without requiring prior knowledge in camera poses or
camera pose distributions. VMRF introduces a view matching scheme, which
exploits unbalanced optimal transport to produce a feature transport plan for
mapping a rendered image with randomly initialized camera pose to the
corresponding real image. With the feature transport plan as the guidance, a
novel pose calibration technique is designed which rectifies the initially
randomized camera poses by predicting relative pose transformations between the
pair of rendered and real images. Extensive experiments over a number of
synthetic and real datasets show that the proposed VMRF outperforms the
state-of-the-art qualitatively and quantitatively by large margins.
|
2209.12065 | Tashreef Muhammad | Nibir Chandra Mandal, Tashreef Muhammad and G. M. Shahariar | Can Transformer Models Effectively Detect Software Aspects in
StackOverflow Discussion? | 15 pages, 2 figures, submitted to International Conference on Machine
Intelligence and Emerging Technologies (MIET 2022) | Preprint of an article published in Machine Intelligence and
Emerging Technologies. MIET 2022. Lecture Notes of the Institute for Computer
Sciences, Social Informatics and Telecommunications Engineering, vol 491.
Springer, Cham | 10.1007/978-3-031-34622-4_18 | null | cs.SE cs.CL | http://creativecommons.org/licenses/by/4.0/ | Dozens of new tools and technologies are being incorporated to help
developers, which is becoming a source of consternation as they struggle to
choose one over the others. For example, there are at least ten frameworks
available to developers for developing web applications, posing a conundrum in
selecting the best one that meets their needs. As a result, developers are
continuously searching for all of the benefits and drawbacks of each API,
framework, tool, and so on. One of the typical approaches is to examine all of
the features through official documentation and discussion. This approach is
time-consuming, often makes it difficult to determine which aspects are the
most important to a particular developer and whether a particular aspect is
important to the community at large. In this paper, we have used a benchmark
API aspects dataset (Opiner) collected from StackOverflow posts and observed
how Transformer models (BERT, RoBERTa, DistilBERT, and XLNet) perform in
detecting software aspects in textual developer discussion with respect to the
baseline Support Vector Machine (SVM) model. Through extensive experimentation,
we have found that transformer models improve the performance of baseline SVM
for most of the aspects, i.e., `Performance', `Security', `Usability',
`Documentation', `Bug', `Legal', `OnlySentiment', and `Others'. However, the
models fail to apprehend some of the aspects (e.g., `Community' and
`Potability') and their performance varies depending on the aspects. Also,
larger architectures like XLNet are ineffective in interpreting software
aspects compared to smaller architectures like DistilBERT.
| [
{
"version": "v1",
"created": "Sat, 24 Sep 2022 18:28:14 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Mandal",
"Nibir Chandra",
""
],
[
"Muhammad",
"Tashreef",
""
],
[
"Shahariar",
"G. M.",
""
]
] | TITLE: Can Transformer Models Effectively Detect Software Aspects in
StackOverflow Discussion?
ABSTRACT: Dozens of new tools and technologies are being incorporated to help
developers, which is becoming a source of consternation as they struggle to
choose one over the others. For example, there are at least ten frameworks
available to developers for developing web applications, posing a conundrum in
selecting the best one that meets their needs. As a result, developers are
continuously searching for all of the benefits and drawbacks of each API,
framework, tool, and so on. One of the typical approaches is to examine all of
the features through official documentation and discussion. This approach is
time-consuming, often makes it difficult to determine which aspects are the
most important to a particular developer and whether a particular aspect is
important to the community at large. In this paper, we have used a benchmark
API aspects dataset (Opiner) collected from StackOverflow posts and observed
how Transformer models (BERT, RoBERTa, DistilBERT, and XLNet) perform in
detecting software aspects in textual developer discussion with respect to the
baseline Support Vector Machine (SVM) model. Through extensive experimentation,
we have found that transformer models improve the performance of baseline SVM
for most of the aspects, i.e., `Performance', `Security', `Usability',
`Documentation', `Bug', `Legal', `OnlySentiment', and `Others'. However, the
models fail to apprehend some of the aspects (e.g., `Community' and
`Potability') and their performance varies depending on the aspects. Also,
larger architectures like XLNet are ineffective in interpreting software
aspects compared to smaller architectures like DistilBERT.
|
2209.15567 | Armita Nourmohammad | Gian Marco Visani, Michael N. Pun, Arman Angaji, Armita Nourmohammad | Holographic-(V)AE: an end-to-end SO(3)-Equivariant (Variational)
Autoencoder in Fourier Space | null | null | null | null | cs.LG physics.bio-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Group-equivariant neural networks have emerged as a data-efficient approach
to solve classification and regression tasks, while respecting the relevant
symmetries of the data. However, little work has been done to extend this
paradigm to the unsupervised and generative domains. Here, we present
Holographic-(Variational) Auto Encoder (H-(V)AE), a fully end-to-end
SO(3)-equivariant (variational) autoencoder in Fourier space, suitable for
unsupervised learning and generation of data distributed around a specified
origin in 3D. H-(V)AE is trained to reconstruct the spherical Fourier encoding
of data, learning in the process a low-dimensional representation of the data
(i.e., a latent space) with a maximally informative rotationally invariant
embedding alongside an equivariant frame describing the orientation of the
data. We extensively test the performance of H-(V)AE on diverse datasets. We
show that the learned latent space efficiently encodes the categorical features
of spherical images. Moreover, H-(V)AE's latent space can be used to extract
compact embeddings for protein structure microenvironments, and when paired
with a Random Forest Regressor, it enables state-of-the-art predictions of
protein-ligand binding affinity.
| [
{
"version": "v1",
"created": "Fri, 30 Sep 2022 16:25:20 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 03:54:09 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Visani",
"Gian Marco",
""
],
[
"Pun",
"Michael N.",
""
],
[
"Angaji",
"Arman",
""
],
[
"Nourmohammad",
"Armita",
""
]
] | TITLE: Holographic-(V)AE: an end-to-end SO(3)-Equivariant (Variational)
Autoencoder in Fourier Space
ABSTRACT: Group-equivariant neural networks have emerged as a data-efficient approach
to solve classification and regression tasks, while respecting the relevant
symmetries of the data. However, little work has been done to extend this
paradigm to the unsupervised and generative domains. Here, we present
Holographic-(Variational) Auto Encoder (H-(V)AE), a fully end-to-end
SO(3)-equivariant (variational) autoencoder in Fourier space, suitable for
unsupervised learning and generation of data distributed around a specified
origin in 3D. H-(V)AE is trained to reconstruct the spherical Fourier encoding
of data, learning in the process a low-dimensional representation of the data
(i.e., a latent space) with a maximally informative rotationally invariant
embedding alongside an equivariant frame describing the orientation of the
data. We extensively test the performance of H-(V)AE on diverse datasets. We
show that the learned latent space efficiently encodes the categorical features
of spherical images. Moreover, H-(V)AE's latent space can be used to extract
compact embeddings for protein structure microenvironments, and when paired
with a Random Forest Regressor, it enables state-of-the-art predictions of
protein-ligand binding affinity.
|
2210.00762 | Jonas Rothfuss | Jonas Rothfuss, Christopher Koenig, Alisa Rupenyan, Andreas Krause | Meta-Learning Priors for Safe Bayesian Optimization | Conference on Robot Learning (CoRL) 2022 | null | null | null | cs.LG cs.AI cs.RO stat.ML | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In robotics, optimizing controller parameters under safety constraints is an
important challenge. Safe Bayesian optimization (BO) quantifies uncertainty in
the objective and constraints to safely guide exploration in such settings.
Hand-designing a suitable probabilistic model can be challenging, however. In
the presence of unknown safety constraints, it is crucial to choose reliable
model hyper-parameters to avoid safety violations. Here, we propose a
data-driven approach to this problem by meta-learning priors for safe BO from
offline data. We build on a meta-learning algorithm, F-PACOH, capable of
providing reliable uncertainty quantification in settings of data scarcity. As
core contribution, we develop a novel framework for choosing safety-compliant
priors in a data-riven manner via empirical uncertainty metrics and a frontier
search algorithm. On benchmark functions and a high-precision motion system, we
demonstrate that our meta-learned priors accelerate the convergence of safe BO
approaches while maintaining safety.
| [
{
"version": "v1",
"created": "Mon, 3 Oct 2022 08:38:38 GMT"
},
{
"version": "v2",
"created": "Fri, 16 Dec 2022 22:51:36 GMT"
},
{
"version": "v3",
"created": "Mon, 12 Jun 2023 14:05:33 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Rothfuss",
"Jonas",
""
],
[
"Koenig",
"Christopher",
""
],
[
"Rupenyan",
"Alisa",
""
],
[
"Krause",
"Andreas",
""
]
] | TITLE: Meta-Learning Priors for Safe Bayesian Optimization
ABSTRACT: In robotics, optimizing controller parameters under safety constraints is an
important challenge. Safe Bayesian optimization (BO) quantifies uncertainty in
the objective and constraints to safely guide exploration in such settings.
Hand-designing a suitable probabilistic model can be challenging, however. In
the presence of unknown safety constraints, it is crucial to choose reliable
model hyper-parameters to avoid safety violations. Here, we propose a
data-driven approach to this problem by meta-learning priors for safe BO from
offline data. We build on a meta-learning algorithm, F-PACOH, capable of
providing reliable uncertainty quantification in settings of data scarcity. As
core contribution, we develop a novel framework for choosing safety-compliant
priors in a data-riven manner via empirical uncertainty metrics and a frontier
search algorithm. On benchmark functions and a high-precision motion system, we
demonstrate that our meta-learned priors accelerate the convergence of safe BO
approaches while maintaining safety.
|
2210.06313 | Chong You | Zonglin Li, Chong You, Srinadh Bhojanapalli, Daliang Li, Ankit Singh
Rawat, Sashank J. Reddi, Ke Ye, Felix Chern, Felix Yu, Ruiqi Guo, Sanjiv
Kumar | The Lazy Neuron Phenomenon: On Emergence of Activation Sparsity in
Transformers | A short version was presented at ICLR 2023. Previous title: Large
Models are Parsimonious Learners: Activation Sparsity in Trained Transformers | null | null | null | cs.LG cs.CL cs.CV stat.ML | http://creativecommons.org/licenses/by/4.0/ | This paper studies the curious phenomenon for machine learning models with
Transformer architectures that their activation maps are sparse. By activation
map we refer to the intermediate output of the multi-layer perceptrons (MLPs)
after a ReLU activation function, and by sparse we mean that on average very
few entries (e.g., 3.0% for T5-Base and 6.3% for ViT-B16) are nonzero for each
input to MLP. Moreover, larger Transformers with more layers and wider MLP
hidden dimensions are sparser as measured by the percentage of nonzero entries.
Through extensive experiments we demonstrate that the emergence of sparsity is
a prevalent phenomenon that occurs for both natural language processing and
vision tasks, on both training and evaluation data, for Transformers of various
configurations, at layers of all depth levels, as well as for other
architectures including MLP-mixers and 2-layer MLPs. We show that sparsity also
emerges using training datasets with random labels, or with random inputs, or
with infinite amount of data, demonstrating that sparsity is not a result of a
specific family of datasets. We discuss how sparsity immediately implies a way
to significantly reduce the FLOP count and improve efficiency for Transformers.
Moreover, we demonstrate perhaps surprisingly that enforcing an even sparser
activation via Top-k thresholding with a small value of k brings a collection
of desired but missing properties for Transformers, namely less sensitivity to
noisy training data, more robustness to input corruptions, and better
calibration for their prediction confidence.
| [
{
"version": "v1",
"created": "Wed, 12 Oct 2022 15:25:19 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 21:53:43 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Li",
"Zonglin",
""
],
[
"You",
"Chong",
""
],
[
"Bhojanapalli",
"Srinadh",
""
],
[
"Li",
"Daliang",
""
],
[
"Rawat",
"Ankit Singh",
""
],
[
"Reddi",
"Sashank J.",
""
],
[
"Ye",
"Ke",
""
],
[
"Chern",
"Felix",
""
],
[
"Yu",
"Felix",
""
],
[
"Guo",
"Ruiqi",
""
],
[
"Kumar",
"Sanjiv",
""
]
] | TITLE: The Lazy Neuron Phenomenon: On Emergence of Activation Sparsity in
Transformers
ABSTRACT: This paper studies the curious phenomenon for machine learning models with
Transformer architectures that their activation maps are sparse. By activation
map we refer to the intermediate output of the multi-layer perceptrons (MLPs)
after a ReLU activation function, and by sparse we mean that on average very
few entries (e.g., 3.0% for T5-Base and 6.3% for ViT-B16) are nonzero for each
input to MLP. Moreover, larger Transformers with more layers and wider MLP
hidden dimensions are sparser as measured by the percentage of nonzero entries.
Through extensive experiments we demonstrate that the emergence of sparsity is
a prevalent phenomenon that occurs for both natural language processing and
vision tasks, on both training and evaluation data, for Transformers of various
configurations, at layers of all depth levels, as well as for other
architectures including MLP-mixers and 2-layer MLPs. We show that sparsity also
emerges using training datasets with random labels, or with random inputs, or
with infinite amount of data, demonstrating that sparsity is not a result of a
specific family of datasets. We discuss how sparsity immediately implies a way
to significantly reduce the FLOP count and improve efficiency for Transformers.
Moreover, we demonstrate perhaps surprisingly that enforcing an even sparser
activation via Top-k thresholding with a small value of k brings a collection
of desired but missing properties for Transformers, namely less sensitivity to
noisy training data, more robustness to input corruptions, and better
calibration for their prediction confidence.
|
2210.06629 | Kishaloy Halder | Siddharth Varia, Shuai Wang, Kishaloy Halder, Robert Vacareanu, Miguel
Ballesteros, Yassine Benajiba, Neha Anna John, Rishita Anubhai, Smaranda
Muresan, Dan Roth | Instruction Tuning for Few-Shot Aspect-Based Sentiment Analysis | Camera ready copy for WASSA at ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Aspect-based Sentiment Analysis (ABSA) is a fine-grained sentiment analysis
task which involves four elements from user-generated texts: aspect term,
aspect category, opinion term, and sentiment polarity. Most computational
approaches focus on some of the ABSA sub-tasks such as tuple (aspect term,
sentiment polarity) or triplet (aspect term, opinion term, sentiment polarity)
extraction using either pipeline or joint modeling approaches. Recently,
generative approaches have been proposed to extract all four elements as (one
or more) quadruplets from text as a single task. In this work, we take a step
further and propose a unified framework for solving ABSA, and the associated
sub-tasks to improve the performance in few-shot scenarios. To this end, we
fine-tune a T5 model with instructional prompts in a multi-task learning
fashion covering all the sub-tasks, as well as the entire quadruple prediction
task. In experiments with multiple benchmark datasets, we show that the
proposed multi-task prompting approach brings performance boost (by absolute
8.29 F1) in the few-shot learning setting.
| [
{
"version": "v1",
"created": "Wed, 12 Oct 2022 23:38:57 GMT"
},
{
"version": "v2",
"created": "Sun, 11 Jun 2023 22:47:33 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Varia",
"Siddharth",
""
],
[
"Wang",
"Shuai",
""
],
[
"Halder",
"Kishaloy",
""
],
[
"Vacareanu",
"Robert",
""
],
[
"Ballesteros",
"Miguel",
""
],
[
"Benajiba",
"Yassine",
""
],
[
"John",
"Neha Anna",
""
],
[
"Anubhai",
"Rishita",
""
],
[
"Muresan",
"Smaranda",
""
],
[
"Roth",
"Dan",
""
]
] | TITLE: Instruction Tuning for Few-Shot Aspect-Based Sentiment Analysis
ABSTRACT: Aspect-based Sentiment Analysis (ABSA) is a fine-grained sentiment analysis
task which involves four elements from user-generated texts: aspect term,
aspect category, opinion term, and sentiment polarity. Most computational
approaches focus on some of the ABSA sub-tasks such as tuple (aspect term,
sentiment polarity) or triplet (aspect term, opinion term, sentiment polarity)
extraction using either pipeline or joint modeling approaches. Recently,
generative approaches have been proposed to extract all four elements as (one
or more) quadruplets from text as a single task. In this work, we take a step
further and propose a unified framework for solving ABSA, and the associated
sub-tasks to improve the performance in few-shot scenarios. To this end, we
fine-tune a T5 model with instructional prompts in a multi-task learning
fashion covering all the sub-tasks, as well as the entire quadruple prediction
task. In experiments with multiple benchmark datasets, we show that the
proposed multi-task prompting approach brings performance boost (by absolute
8.29 F1) in the few-shot learning setting.
|
2210.13319 | Krunoslav Lehman Pavasovic | Krunoslav Lehman Pavasovic, Jonas Rothfuss and Andreas Krause | MARS: Meta-Learning as Score Matching in the Function Space | In International Conference on Learning Representations (ICLR), 2023 | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Meta-learning aims to extract useful inductive biases from a set of related
datasets. In Bayesian meta-learning, this is typically achieved by constructing
a prior distribution over neural network parameters. However, specifying
families of computationally viable prior distributions over the
high-dimensional neural network parameters is difficult. As a result, existing
approaches resort to meta-learning restrictive diagonal Gaussian priors,
severely limiting their expressiveness and performance. To circumvent these
issues, we approach meta-learning through the lens of functional Bayesian
neural network inference, which views the prior as a stochastic process and
performs inference in the function space. Specifically, we view the
meta-training tasks as samples from the data-generating process and formalize
meta-learning as empirically estimating the law of this stochastic process. Our
approach can seamlessly acquire and represent complex prior knowledge by
meta-learning the score function of the data-generating process marginals
instead of parameter space priors. In a comprehensive benchmark, we demonstrate
that our method achieves state-of-the-art performance in terms of predictive
accuracy and substantial improvements in the quality of uncertainty estimates.
| [
{
"version": "v1",
"created": "Mon, 24 Oct 2022 15:14:26 GMT"
},
{
"version": "v2",
"created": "Mon, 20 Feb 2023 09:59:14 GMT"
},
{
"version": "v3",
"created": "Sat, 10 Jun 2023 10:11:33 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Pavasovic",
"Krunoslav Lehman",
""
],
[
"Rothfuss",
"Jonas",
""
],
[
"Krause",
"Andreas",
""
]
] | TITLE: MARS: Meta-Learning as Score Matching in the Function Space
ABSTRACT: Meta-learning aims to extract useful inductive biases from a set of related
datasets. In Bayesian meta-learning, this is typically achieved by constructing
a prior distribution over neural network parameters. However, specifying
families of computationally viable prior distributions over the
high-dimensional neural network parameters is difficult. As a result, existing
approaches resort to meta-learning restrictive diagonal Gaussian priors,
severely limiting their expressiveness and performance. To circumvent these
issues, we approach meta-learning through the lens of functional Bayesian
neural network inference, which views the prior as a stochastic process and
performs inference in the function space. Specifically, we view the
meta-training tasks as samples from the data-generating process and formalize
meta-learning as empirically estimating the law of this stochastic process. Our
approach can seamlessly acquire and represent complex prior knowledge by
meta-learning the score function of the data-generating process marginals
instead of parameter space priors. In a comprehensive benchmark, we demonstrate
that our method achieves state-of-the-art performance in terms of predictive
accuracy and substantial improvements in the quality of uncertainty estimates.
|
2211.09146 | Benjia Zhou | Benjia Zhou, Pichao Wang, Jun Wan, Yanyan Liang and Fan Wang | A Unified Multimodal De- and Re-coupling Framework for RGB-D Motion
Recognition | Accepted to TPAMI 2023 | null | null | null | cs.CV cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Motion recognition is a promising direction in computer vision, but the
training of video classification models is much harder than images due to
insufficient data and considerable parameters. To get around this, some works
strive to explore multimodal cues from RGB-D data. Although improving motion
recognition to some extent, these methods still face sub-optimal situations in
the following aspects: (i) Data augmentation, i.e., the scale of the RGB-D
datasets is still limited, and few efforts have been made to explore novel data
augmentation strategies for videos; (ii) Optimization mechanism, i.e., the
tightly space-time-entangled network structure brings more challenges to
spatiotemporal information modeling; And (iii) cross-modal knowledge fusion,
i.e., the high similarity between multimodal representations caused to
insufficient late fusion. To alleviate these drawbacks, we propose to improve
RGB-D-based motion recognition both from data and algorithm perspectives in
this paper. In more detail, firstly, we introduce a novel video data
augmentation method dubbed ShuffleMix, which acts as a supplement to MixUp, to
provide additional temporal regularization for motion recognition. Secondly, a
Unified Multimodal De-coupling and multi-stage Re-coupling framework, termed
UMDR, is proposed for video representation learning. Finally, a novel
cross-modal Complement Feature Catcher (CFCer) is explored to mine potential
commonalities features in multimodal information as the auxiliary fusion
stream, to improve the late fusion results. The seamless combination of these
novel designs forms a robust spatiotemporal representation and achieves better
performance than state-of-the-art methods on four public motion datasets.
Specifically, UMDR achieves unprecedented improvements of +4.5% on the Chalearn
IsoGD dataset. Our code is available at
https://github.com/zhoubenjia/MotionRGBD-PAMI.
| [
{
"version": "v1",
"created": "Wed, 16 Nov 2022 19:00:23 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 08:47:22 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zhou",
"Benjia",
""
],
[
"Wang",
"Pichao",
""
],
[
"Wan",
"Jun",
""
],
[
"Liang",
"Yanyan",
""
],
[
"Wang",
"Fan",
""
]
] | TITLE: A Unified Multimodal De- and Re-coupling Framework for RGB-D Motion
Recognition
ABSTRACT: Motion recognition is a promising direction in computer vision, but the
training of video classification models is much harder than images due to
insufficient data and considerable parameters. To get around this, some works
strive to explore multimodal cues from RGB-D data. Although improving motion
recognition to some extent, these methods still face sub-optimal situations in
the following aspects: (i) Data augmentation, i.e., the scale of the RGB-D
datasets is still limited, and few efforts have been made to explore novel data
augmentation strategies for videos; (ii) Optimization mechanism, i.e., the
tightly space-time-entangled network structure brings more challenges to
spatiotemporal information modeling; And (iii) cross-modal knowledge fusion,
i.e., the high similarity between multimodal representations caused to
insufficient late fusion. To alleviate these drawbacks, we propose to improve
RGB-D-based motion recognition both from data and algorithm perspectives in
this paper. In more detail, firstly, we introduce a novel video data
augmentation method dubbed ShuffleMix, which acts as a supplement to MixUp, to
provide additional temporal regularization for motion recognition. Secondly, a
Unified Multimodal De-coupling and multi-stage Re-coupling framework, termed
UMDR, is proposed for video representation learning. Finally, a novel
cross-modal Complement Feature Catcher (CFCer) is explored to mine potential
commonalities features in multimodal information as the auxiliary fusion
stream, to improve the late fusion results. The seamless combination of these
novel designs forms a robust spatiotemporal representation and achieves better
performance than state-of-the-art methods on four public motion datasets.
Specifically, UMDR achieves unprecedented improvements of +4.5% on the Chalearn
IsoGD dataset. Our code is available at
https://github.com/zhoubenjia/MotionRGBD-PAMI.
|
2211.10934 | Akira Taniguchi | Akira Taniguchi, Yoshiki Tabuchi, Tomochika Ishikawa, Lotfi El Hafi,
Yoshinobu Hagiwara, Tadahiro Taniguchi | Active Exploration based on Information Gain by Particle Filter for
Efficient Spatial Concept Formation | Accepted to Advanced Robotics | null | 10.1080/01691864.2023.2225175 | null | cs.RO cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Autonomous robots need to learn the categories of various places by exploring
their environments and interacting with users. However, preparing training
datasets with linguistic instructions from users is time-consuming and
labor-intensive. Moreover, effective exploration is essential for appropriate
concept formation and rapid environmental coverage. To address this issue, we
propose an active inference method, referred to as spatial concept formation
with information gain-based active exploration (SpCoAE) that combines
sequential Bayesian inference using particle filters and information gain-based
destination determination in a probabilistic generative model. This study
interprets the robot's action as a selection of destinations to ask the user,
`What kind of place is this?' in the context of active inference. This study
provides insights into the technical aspects of the proposed method, including
active perception and exploration by the robot, and how the method can enable
mobile robots to learn spatial concepts through active exploration. Our
experiment demonstrated the effectiveness of the SpCoAE in efficiently
determining a destination for learning appropriate spatial concepts in home
environments.
| [
{
"version": "v1",
"created": "Sun, 20 Nov 2022 10:12:17 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 06:10:13 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Taniguchi",
"Akira",
""
],
[
"Tabuchi",
"Yoshiki",
""
],
[
"Ishikawa",
"Tomochika",
""
],
[
"Hafi",
"Lotfi El",
""
],
[
"Hagiwara",
"Yoshinobu",
""
],
[
"Taniguchi",
"Tadahiro",
""
]
] | TITLE: Active Exploration based on Information Gain by Particle Filter for
Efficient Spatial Concept Formation
ABSTRACT: Autonomous robots need to learn the categories of various places by exploring
their environments and interacting with users. However, preparing training
datasets with linguistic instructions from users is time-consuming and
labor-intensive. Moreover, effective exploration is essential for appropriate
concept formation and rapid environmental coverage. To address this issue, we
propose an active inference method, referred to as spatial concept formation
with information gain-based active exploration (SpCoAE) that combines
sequential Bayesian inference using particle filters and information gain-based
destination determination in a probabilistic generative model. This study
interprets the robot's action as a selection of destinations to ask the user,
`What kind of place is this?' in the context of active inference. This study
provides insights into the technical aspects of the proposed method, including
active perception and exploration by the robot, and how the method can enable
mobile robots to learn spatial concepts through active exploration. Our
experiment demonstrated the effectiveness of the SpCoAE in efficiently
determining a destination for learning appropriate spatial concepts in home
environments.
|
2211.11074 | Yehuda Dar | Yehuda Dar, Lorenzo Luzi, Richard G. Baraniuk | Frozen Overparameterization: A Double Descent Perspective on Transfer
Learning of Deep Neural Networks | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study the generalization behavior of transfer learning of deep neural
networks (DNNs). We adopt the overparameterization perspective -- featuring
interpolation of the training data (i.e., approximately zero train error) and
the double descent phenomenon -- to explain the delicate effect of the transfer
learning setting on generalization performance. We study how the generalization
behavior of transfer learning is affected by the dataset size in the source and
target tasks, the number of transferred layers that are kept frozen in the
target DNN training, and the similarity between the source and target tasks. We
show that the test error evolution during the target DNN training has a more
significant double descent effect when the target training dataset is
sufficiently large. In addition, a larger source training dataset can yield a
slower target DNN training. Moreover, we demonstrate that the number of frozen
layers can determine whether the transfer learning is effectively
underparameterized or overparameterized and, in turn, this may induce a
freezing-wise double descent phenomenon that determines the relative success or
failure of learning. Also, we show that the double descent phenomenon may make
a transfer from a less related source task better than a transfer from a more
related source task. We establish our results using image classification
experiments with the ResNet, DenseNet and the vision transformer (ViT)
architectures.
| [
{
"version": "v1",
"created": "Sun, 20 Nov 2022 20:26:23 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 17:39:15 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Dar",
"Yehuda",
""
],
[
"Luzi",
"Lorenzo",
""
],
[
"Baraniuk",
"Richard G.",
""
]
] | TITLE: Frozen Overparameterization: A Double Descent Perspective on Transfer
Learning of Deep Neural Networks
ABSTRACT: We study the generalization behavior of transfer learning of deep neural
networks (DNNs). We adopt the overparameterization perspective -- featuring
interpolation of the training data (i.e., approximately zero train error) and
the double descent phenomenon -- to explain the delicate effect of the transfer
learning setting on generalization performance. We study how the generalization
behavior of transfer learning is affected by the dataset size in the source and
target tasks, the number of transferred layers that are kept frozen in the
target DNN training, and the similarity between the source and target tasks. We
show that the test error evolution during the target DNN training has a more
significant double descent effect when the target training dataset is
sufficiently large. In addition, a larger source training dataset can yield a
slower target DNN training. Moreover, we demonstrate that the number of frozen
layers can determine whether the transfer learning is effectively
underparameterized or overparameterized and, in turn, this may induce a
freezing-wise double descent phenomenon that determines the relative success or
failure of learning. Also, we show that the double descent phenomenon may make
a transfer from a less related source task better than a transfer from a more
related source task. We establish our results using image classification
experiments with the ResNet, DenseNet and the vision transformer (ViT)
architectures.
|
2212.03145 | Shibo Jie | Shibo Jie, Zhi-Hong Deng | FacT: Factor-Tuning for Lightweight Adaptation on Vision Transformer | AAAI 2023 Oral. Code: https://github.com/JieShibo/PETL-ViT | null | null | null | cs.CV | http://creativecommons.org/publicdomain/zero/1.0/ | Recent work has explored the potential to adapt a pre-trained vision
transformer (ViT) by updating only a few parameters so as to improve storage
efficiency, called parameter-efficient transfer learning (PETL). Current PETL
methods have shown that by tuning only 0.5% of the parameters, ViT can be
adapted to downstream tasks with even better performance than full fine-tuning.
In this paper, we aim to further promote the efficiency of PETL to meet the
extreme storage constraint in real-world applications. To this end, we propose
a tensorization-decomposition framework to store the weight increments, in
which the weights of each ViT are tensorized into a single 3D tensor, and their
increments are then decomposed into lightweight factors. In the fine-tuning
process, only the factors need to be updated and stored, termed Factor-Tuning
(FacT). On VTAB-1K benchmark, our method performs on par with NOAH, the
state-of-the-art PETL method, while being 5x more parameter-efficient. We also
present a tiny version that only uses 8K (0.01% of ViT's parameters) trainable
parameters but outperforms full fine-tuning and many other PETL methods such as
VPT and BitFit. In few-shot settings, FacT also beats all PETL baselines using
the fewest parameters, demonstrating its strong capability in the low-data
regime.
| [
{
"version": "v1",
"created": "Tue, 6 Dec 2022 17:18:33 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 08:20:10 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Jie",
"Shibo",
""
],
[
"Deng",
"Zhi-Hong",
""
]
] | TITLE: FacT: Factor-Tuning for Lightweight Adaptation on Vision Transformer
ABSTRACT: Recent work has explored the potential to adapt a pre-trained vision
transformer (ViT) by updating only a few parameters so as to improve storage
efficiency, called parameter-efficient transfer learning (PETL). Current PETL
methods have shown that by tuning only 0.5% of the parameters, ViT can be
adapted to downstream tasks with even better performance than full fine-tuning.
In this paper, we aim to further promote the efficiency of PETL to meet the
extreme storage constraint in real-world applications. To this end, we propose
a tensorization-decomposition framework to store the weight increments, in
which the weights of each ViT are tensorized into a single 3D tensor, and their
increments are then decomposed into lightweight factors. In the fine-tuning
process, only the factors need to be updated and stored, termed Factor-Tuning
(FacT). On VTAB-1K benchmark, our method performs on par with NOAH, the
state-of-the-art PETL method, while being 5x more parameter-efficient. We also
present a tiny version that only uses 8K (0.01% of ViT's parameters) trainable
parameters but outperforms full fine-tuning and many other PETL methods such as
VPT and BitFit. In few-shot settings, FacT also beats all PETL baselines using
the fewest parameters, demonstrating its strong capability in the low-data
regime.
|
2212.07347 | Raghav Kansal | Zichun Hao, Raghav Kansal, Javier Duarte, Nadezda Chernyavskaya | Lorentz group equivariant autoencoders | 11 pages, 7 figures, 4 tables, and a 3 page appendix | Eur. Phys. J. C 83, 485 (2023) | 10.1140/epjc/s10052-023-11633-5 | null | hep-ex cs.LG | http://creativecommons.org/licenses/by/4.0/ | There has been significant work recently in developing machine learning (ML)
models in high energy physics (HEP) for tasks such as classification,
simulation, and anomaly detection. Often these models are adapted from those
designed for datasets in computer vision or natural language processing, which
lack inductive biases suited to HEP data, such as equivariance to its inherent
symmetries. Such biases have been shown to make models more performant and
interpretable, and reduce the amount of training data needed. To that end, we
develop the Lorentz group autoencoder (LGAE), an autoencoder model equivariant
with respect to the proper, orthochronous Lorentz group $\mathrm{SO}^+(3,1)$,
with a latent space living in the representations of the group. We present our
architecture and several experimental results on jets at the LHC and find it
outperforms graph and convolutional neural network baseline models on several
compression, reconstruction, and anomaly detection metrics. We also demonstrate
the advantage of such an equivariant model in analyzing the latent space of the
autoencoder, which can improve the explainability of potential anomalies
discovered by such ML models.
| [
{
"version": "v1",
"created": "Wed, 14 Dec 2022 17:19:46 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 23:28:30 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Hao",
"Zichun",
""
],
[
"Kansal",
"Raghav",
""
],
[
"Duarte",
"Javier",
""
],
[
"Chernyavskaya",
"Nadezda",
""
]
] | TITLE: Lorentz group equivariant autoencoders
ABSTRACT: There has been significant work recently in developing machine learning (ML)
models in high energy physics (HEP) for tasks such as classification,
simulation, and anomaly detection. Often these models are adapted from those
designed for datasets in computer vision or natural language processing, which
lack inductive biases suited to HEP data, such as equivariance to its inherent
symmetries. Such biases have been shown to make models more performant and
interpretable, and reduce the amount of training data needed. To that end, we
develop the Lorentz group autoencoder (LGAE), an autoencoder model equivariant
with respect to the proper, orthochronous Lorentz group $\mathrm{SO}^+(3,1)$,
with a latent space living in the representations of the group. We present our
architecture and several experimental results on jets at the LHC and find it
outperforms graph and convolutional neural network baseline models on several
compression, reconstruction, and anomaly detection metrics. We also demonstrate
the advantage of such an equivariant model in analyzing the latent space of the
autoencoder, which can improve the explainability of potential anomalies
discovered by such ML models.
|
2212.09097 | Peng Li | Yuanchi Zhang, Peng Li, Maosong Sun, Yang Liu | Continual Knowledge Distillation for Neural Machine Translation | Accepted to ACL 2023. The source code is available at
https://github.com/THUNLP-MT/CKD | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While many parallel corpora are not publicly accessible for data copyright,
data privacy and competitive differentiation reasons, trained translation
models are increasingly available on open platforms. In this work, we propose a
method called continual knowledge distillation to take advantage of existing
translation models to improve one model of interest. The basic idea is to
sequentially transfer knowledge from each trained model to the distilled model.
Extensive experiments on Chinese-English and German-English datasets show that
our method achieves significant and consistent improvements over strong
baselines under both homogeneous and heterogeneous trained model settings and
is robust to malicious models.
| [
{
"version": "v1",
"created": "Sun, 18 Dec 2022 14:41:13 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 12:00:20 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zhang",
"Yuanchi",
""
],
[
"Li",
"Peng",
""
],
[
"Sun",
"Maosong",
""
],
[
"Liu",
"Yang",
""
]
] | TITLE: Continual Knowledge Distillation for Neural Machine Translation
ABSTRACT: While many parallel corpora are not publicly accessible for data copyright,
data privacy and competitive differentiation reasons, trained translation
models are increasingly available on open platforms. In this work, we propose a
method called continual knowledge distillation to take advantage of existing
translation models to improve one model of interest. The basic idea is to
sequentially transfer knowledge from each trained model to the distilled model.
Extensive experiments on Chinese-English and German-English datasets show that
our method achieves significant and consistent improvements over strong
baselines under both homogeneous and heterogeneous trained model settings and
is robust to malicious models.
|
2212.10563 | Hadas Orgad | Hadas Orgad, Yonatan Belinkov | BLIND: Bias Removal With No Demographics | Accepted to ACL 2023 main conference | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Models trained on real-world data tend to imitate and amplify social biases.
Common methods to mitigate biases require prior information on the types of
biases that should be mitigated (e.g., gender or racial bias) and the social
groups associated with each data sample. In this work, we introduce BLIND, a
method for bias removal with no prior knowledge of the demographics in the
dataset. While training a model on a downstream task, BLIND detects biased
samples using an auxiliary model that predicts the main model's success, and
down-weights those samples during the training process. Experiments with racial
and gender biases in sentiment classification and occupation classification
tasks demonstrate that BLIND mitigates social biases without relying on a
costly demographic annotation process. Our method is competitive with other
methods that require demographic information and sometimes even surpasses them.
| [
{
"version": "v1",
"created": "Tue, 20 Dec 2022 18:59:42 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 03:55:12 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Orgad",
"Hadas",
""
],
[
"Belinkov",
"Yonatan",
""
]
] | TITLE: BLIND: Bias Removal With No Demographics
ABSTRACT: Models trained on real-world data tend to imitate and amplify social biases.
Common methods to mitigate biases require prior information on the types of
biases that should be mitigated (e.g., gender or racial bias) and the social
groups associated with each data sample. In this work, we introduce BLIND, a
method for bias removal with no prior knowledge of the demographics in the
dataset. While training a model on a downstream task, BLIND detects biased
samples using an auxiliary model that predicts the main model's success, and
down-weights those samples during the training process. Experiments with racial
and gender biases in sentiment classification and occupation classification
tasks demonstrate that BLIND mitigates social biases without relying on a
costly demographic annotation process. Our method is competitive with other
methods that require demographic information and sometimes even surpasses them.
|
2212.10773 | Ying Shen | Zhiyang Xu, Ying Shen, Lifu Huang | MultiInstruct: Improving Multi-Modal Zero-Shot Learning via Instruction
Tuning | ACL 2023, dataset url: https://github.com/VT-NLP/MultiInstruct | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Instruction tuning, a new learning paradigm that fine-tunes pre-trained
language models on tasks specified through instructions, has shown promising
zero-shot performance on various natural language processing tasks. However, it
has yet to be explored for vision and multimodal tasks. In this work, we
introduce MUL-TIINSTRUCT, the first multimodal instruction tuning benchmark
dataset that consists of 62 diverse multimodal tasks in a unified seq-to-seq
format covering 10 broad categories. The tasks are derived from 21 existing
open-source datasets and each task is equipped with 5 expert-written
instructions. We take OFA as the base pre-trained model for multimodal
instruction tuning, and to further improve its zero-shot performance, we
explore multiple transfer learning strategies to leverage the large-scale
NATURAL INSTRUCTIONS dataset. Experimental results demonstrate strong zero-shot
performance on various unseen multimodal tasks and the benefit of transfer
learning from a text-only instruction dataset. We also design a new evaluation
metric - Sensitivity, to evaluate how sensitive the model is to the variety of
instructions. Our results indicate that fine-tuning the model on a diverse set
of tasks and instructions leads to a reduced sensitivity to variations in
instructions for each task.
| [
{
"version": "v1",
"created": "Wed, 21 Dec 2022 05:17:06 GMT"
},
{
"version": "v2",
"created": "Wed, 7 Jun 2023 23:05:26 GMT"
},
{
"version": "v3",
"created": "Sat, 10 Jun 2023 18:33:21 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Xu",
"Zhiyang",
""
],
[
"Shen",
"Ying",
""
],
[
"Huang",
"Lifu",
""
]
] | TITLE: MultiInstruct: Improving Multi-Modal Zero-Shot Learning via Instruction
Tuning
ABSTRACT: Instruction tuning, a new learning paradigm that fine-tunes pre-trained
language models on tasks specified through instructions, has shown promising
zero-shot performance on various natural language processing tasks. However, it
has yet to be explored for vision and multimodal tasks. In this work, we
introduce MUL-TIINSTRUCT, the first multimodal instruction tuning benchmark
dataset that consists of 62 diverse multimodal tasks in a unified seq-to-seq
format covering 10 broad categories. The tasks are derived from 21 existing
open-source datasets and each task is equipped with 5 expert-written
instructions. We take OFA as the base pre-trained model for multimodal
instruction tuning, and to further improve its zero-shot performance, we
explore multiple transfer learning strategies to leverage the large-scale
NATURAL INSTRUCTIONS dataset. Experimental results demonstrate strong zero-shot
performance on various unseen multimodal tasks and the benefit of transfer
learning from a text-only instruction dataset. We also design a new evaluation
metric - Sensitivity, to evaluate how sensitive the model is to the variety of
instructions. Our results indicate that fine-tuning the model on a diverse set
of tasks and instructions leads to a reduced sensitivity to variations in
instructions for each task.
|
2301.00391 | Chunyang Wang | Chunyang Wang, Desen Sun, Yuebin Bai | PiPAD: Pipelined and Parallel Dynamic GNN Training on GPUs | To appear at PPoPP 2023: 28th ACM SIGPLAN Annual Symposium on
Principles and Practice of Parallel Programming | null | null | null | cs.LG cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Dynamic Graph Neural Networks (DGNNs) have been broadly applied in various
real-life applications, such as link prediction and pandemic forecast, to
capture both static structural information and temporal characteristics from
dynamic graphs. Combining both time-dependent and -independent components,
DGNNs manifest substantial parallel computation and data reuse potentials, but
suffer from severe memory access inefficiency and data transfer overhead under
the canonical one-graph-at-a-time training pattern. To tackle the challenges,
we propose PiPAD, a $\underline{\textbf{Pi}}pelined$ and
$\underline{\textbf{PA}}rallel$ $\underline{\textbf{D}}GNN$ training framework
for the end-to-end performance optimization on GPUs. From both the algorithm
and runtime level, PiPAD holistically reconstructs the overall training
paradigm from the data organization to computation manner. Capable of
processing multiple graph snapshots in parallel, PiPAD eliminates the
unnecessary data transmission and alleviates memory access inefficiency to
improve the overall performance. Our evaluation across various datasets shows
PiPAD achieves $1.22\times$-$9.57\times$ speedup over the state-of-the-art DGNN
frameworks on three representative models.
| [
{
"version": "v1",
"created": "Sun, 1 Jan 2023 12:10:31 GMT"
},
{
"version": "v2",
"created": "Thu, 5 Jan 2023 11:30:22 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Wang",
"Chunyang",
""
],
[
"Sun",
"Desen",
""
],
[
"Bai",
"Yuebin",
""
]
] | TITLE: PiPAD: Pipelined and Parallel Dynamic GNN Training on GPUs
ABSTRACT: Dynamic Graph Neural Networks (DGNNs) have been broadly applied in various
real-life applications, such as link prediction and pandemic forecast, to
capture both static structural information and temporal characteristics from
dynamic graphs. Combining both time-dependent and -independent components,
DGNNs manifest substantial parallel computation and data reuse potentials, but
suffer from severe memory access inefficiency and data transfer overhead under
the canonical one-graph-at-a-time training pattern. To tackle the challenges,
we propose PiPAD, a $\underline{\textbf{Pi}}pelined$ and
$\underline{\textbf{PA}}rallel$ $\underline{\textbf{D}}GNN$ training framework
for the end-to-end performance optimization on GPUs. From both the algorithm
and runtime level, PiPAD holistically reconstructs the overall training
paradigm from the data organization to computation manner. Capable of
processing multiple graph snapshots in parallel, PiPAD eliminates the
unnecessary data transmission and alleviates memory access inefficiency to
improve the overall performance. Our evaluation across various datasets shows
PiPAD achieves $1.22\times$-$9.57\times$ speedup over the state-of-the-art DGNN
frameworks on three representative models.
|
2301.00488 | Suayb Arslan | Suayb S. Arslan and Pawan Sinha | Information Transfer Rate in BCIs: Towards Tightly Integrated Symbiosis | 27 pages, 10 figures, 2 tables. Accepted to Biomedical Signal
Processing and Control, Elsevier, 2023 | null | null | null | cs.HC cs.AI cs.IT cs.PF eess.SP math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The information transmission rate (ITR), or effective bit rate, is a popular
and widely used information measurement metric, particularly popularized for
SSVEP-based Brain-Computer (BCI) interfaces. By combining speed and accuracy
into a single-valued parameter, this metric aids in the evaluation and
comparison of various target identification algorithms across different BCI
communities. In order to calculate ITR, it is customary to assume a uniform
input distribution and an oversimplified channel model that is memoryless,
stationary, and symmetrical in nature with discrete alphabet sizes. To
accurately depict performance and inspire an end-to-end design for futuristic
BCI designs, a more thorough examination and definition of ITR is therefore
required. We model the symbiotic communication medium, hosted by the
retinogeniculate visual pathway, as a discrete memoryless channel and use the
modified capacity expressions to redefine the ITR. We leverage a result for
directed graphs to characterize the relationship between the asymmetry of the
transition statistics and the ITR gain due to the new definition, leading to
potential bounds on data rate performance. On two well-known SSVEP datasets, we
compared two cutting-edge target identification methods. Results indicate that
the induced DM channel asymmetry has a greater impact on the actual perceived
ITR than the change in input distribution. Moreover, it is demonstrated that
the ITR gain under the new definition is inversely correlated with the
asymmetry in the channel transition statistics. Individual input customizations
are further shown to yield perceived ITR performance improvements. Finally, an
algorithm is proposed to find the capacity of binary classification and further
discussions are given to extend such results to multi-class case through
ensemble techniques.
| [
{
"version": "v1",
"created": "Sun, 1 Jan 2023 23:07:59 GMT"
},
{
"version": "v2",
"created": "Tue, 3 Jan 2023 02:00:51 GMT"
},
{
"version": "v3",
"created": "Sun, 11 Jun 2023 02:59:39 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Arslan",
"Suayb S.",
""
],
[
"Sinha",
"Pawan",
""
]
] | TITLE: Information Transfer Rate in BCIs: Towards Tightly Integrated Symbiosis
ABSTRACT: The information transmission rate (ITR), or effective bit rate, is a popular
and widely used information measurement metric, particularly popularized for
SSVEP-based Brain-Computer (BCI) interfaces. By combining speed and accuracy
into a single-valued parameter, this metric aids in the evaluation and
comparison of various target identification algorithms across different BCI
communities. In order to calculate ITR, it is customary to assume a uniform
input distribution and an oversimplified channel model that is memoryless,
stationary, and symmetrical in nature with discrete alphabet sizes. To
accurately depict performance and inspire an end-to-end design for futuristic
BCI designs, a more thorough examination and definition of ITR is therefore
required. We model the symbiotic communication medium, hosted by the
retinogeniculate visual pathway, as a discrete memoryless channel and use the
modified capacity expressions to redefine the ITR. We leverage a result for
directed graphs to characterize the relationship between the asymmetry of the
transition statistics and the ITR gain due to the new definition, leading to
potential bounds on data rate performance. On two well-known SSVEP datasets, we
compared two cutting-edge target identification methods. Results indicate that
the induced DM channel asymmetry has a greater impact on the actual perceived
ITR than the change in input distribution. Moreover, it is demonstrated that
the ITR gain under the new definition is inversely correlated with the
asymmetry in the channel transition statistics. Individual input customizations
are further shown to yield perceived ITR performance improvements. Finally, an
algorithm is proposed to find the capacity of binary classification and further
discussions are given to extend such results to multi-class case through
ensemble techniques.
|
2301.01919 | Xudong Guo | Xudong Guo, Daming Shi, Wenhui Fan | Scalable Communication for Multi-Agent Reinforcement Learning via
Transformer-Based Email Mechanism | Accepted by IJCAI 2023 | null | null | null | cs.MA cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Communication can impressively improve cooperation in multi-agent
reinforcement learning (MARL), especially for partially-observed tasks.
However, existing works either broadcast the messages leading to information
redundancy, or learn targeted communication by modeling all the other agents as
targets, which is not scalable when the number of agents varies. In this work,
to tackle the scalability problem of MARL communication for partially-observed
tasks, we propose a novel framework Transformer-based Email Mechanism (TEM).
The agents adopt local communication to send messages only to the ones that can
be observed without modeling all the agents. Inspired by human cooperation with
email forwarding, we design message chains to forward information to cooperate
with the agents outside the observation range. We introduce Transformer to
encode and decode the message chain to choose the next receiver selectively.
Empirically, TEM outperforms the baselines on multiple cooperative MARL
benchmarks. When the number of agents varies, TEM maintains superior
performance without further training.
| [
{
"version": "v1",
"created": "Thu, 5 Jan 2023 05:34:30 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 07:13:41 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Guo",
"Xudong",
""
],
[
"Shi",
"Daming",
""
],
[
"Fan",
"Wenhui",
""
]
] | TITLE: Scalable Communication for Multi-Agent Reinforcement Learning via
Transformer-Based Email Mechanism
ABSTRACT: Communication can impressively improve cooperation in multi-agent
reinforcement learning (MARL), especially for partially-observed tasks.
However, existing works either broadcast the messages leading to information
redundancy, or learn targeted communication by modeling all the other agents as
targets, which is not scalable when the number of agents varies. In this work,
to tackle the scalability problem of MARL communication for partially-observed
tasks, we propose a novel framework Transformer-based Email Mechanism (TEM).
The agents adopt local communication to send messages only to the ones that can
be observed without modeling all the agents. Inspired by human cooperation with
email forwarding, we design message chains to forward information to cooperate
with the agents outside the observation range. We introduce Transformer to
encode and decode the message chain to choose the next receiver selectively.
Empirically, TEM outperforms the baselines on multiple cooperative MARL
benchmarks. When the number of agents varies, TEM maintains superior
performance without further training.
|
2301.03426 | Ibrahim Hroob | Ibrahim Hroob, Sergi Molina, Riccardo Polvara, Grzegorz Cielniak and
Marc Hanheide | LTS-NET: End-to-end Unsupervised Learning of Long-Term 3D Stable objects | null | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this research, we present an end-to-end data-driven pipeline for
determining the long-term stability status of objects within a given
environment, specifically distinguishing between static and dynamic objects.
Understanding object stability is key for mobile robots since long-term stable
objects can be exploited as landmarks for long-term localisation. Our pipeline
includes a labelling method that utilizes historical data from the environment
to generate training data for a neural network. Rather than utilizing discrete
labels, we propose the use of point-wise continuous label values, indicating
the spatio-temporal stability of individual points, to train a point cloud
regression network named LTS-NET. Our approach is evaluated on point cloud data
from two parking lots in the NCLT dataset, and the results show that our
proposed solution, outperforms direct training of a classification model for
static vs dynamic object classification.
| [
{
"version": "v1",
"created": "Mon, 9 Jan 2023 15:24:19 GMT"
},
{
"version": "v2",
"created": "Thu, 26 Jan 2023 05:04:37 GMT"
},
{
"version": "v3",
"created": "Mon, 12 Jun 2023 07:18:39 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Hroob",
"Ibrahim",
""
],
[
"Molina",
"Sergi",
""
],
[
"Polvara",
"Riccardo",
""
],
[
"Cielniak",
"Grzegorz",
""
],
[
"Hanheide",
"Marc",
""
]
] | TITLE: LTS-NET: End-to-end Unsupervised Learning of Long-Term 3D Stable objects
ABSTRACT: In this research, we present an end-to-end data-driven pipeline for
determining the long-term stability status of objects within a given
environment, specifically distinguishing between static and dynamic objects.
Understanding object stability is key for mobile robots since long-term stable
objects can be exploited as landmarks for long-term localisation. Our pipeline
includes a labelling method that utilizes historical data from the environment
to generate training data for a neural network. Rather than utilizing discrete
labels, we propose the use of point-wise continuous label values, indicating
the spatio-temporal stability of individual points, to train a point cloud
regression network named LTS-NET. Our approach is evaluated on point cloud data
from two parking lots in the NCLT dataset, and the results show that our
proposed solution, outperforms direct training of a classification model for
static vs dynamic object classification.
|
2301.05119 | Francesco Pierri | Francesco Pierri, Geng Liu, Stefano Ceri | ITA-ELECTION-2022: A multi-platform dataset of social media
conversations around the 2022 Italian general election | 4 pages, 3 figures, 2 tables | null | null | null | cs.SI | http://creativecommons.org/licenses/by-sa/4.0/ | Online social media play a major role in shaping public discourse and
opinion, especially during political events. We present the first public
multi-platform dataset of Italian-language political conversations, focused on
the 2022 Italian general election taking place on September 25th. Leveraging
public APIs and a keyword-based search, we collected millions of posts
published by users, pages and groups on Facebook, Instagram and Twitter, along
with metadata of TikTok and YouTube videos shared on these platforms, over a
period of four months. We augmented the dataset with a collection of political
ads sponsored on Meta platforms, and a list of social media handles associated
with political representatives. Our data resource will allow researchers and
academics to further our understanding of the role of social media in the
democratic process.
| [
{
"version": "v1",
"created": "Thu, 12 Jan 2023 16:19:08 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 10:17:54 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Pierri",
"Francesco",
""
],
[
"Liu",
"Geng",
""
],
[
"Ceri",
"Stefano",
""
]
] | TITLE: ITA-ELECTION-2022: A multi-platform dataset of social media
conversations around the 2022 Italian general election
ABSTRACT: Online social media play a major role in shaping public discourse and
opinion, especially during political events. We present the first public
multi-platform dataset of Italian-language political conversations, focused on
the 2022 Italian general election taking place on September 25th. Leveraging
public APIs and a keyword-based search, we collected millions of posts
published by users, pages and groups on Facebook, Instagram and Twitter, along
with metadata of TikTok and YouTube videos shared on these platforms, over a
period of four months. We augmented the dataset with a collection of political
ads sponsored on Meta platforms, and a list of social media handles associated
with political representatives. Our data resource will allow researchers and
academics to further our understanding of the role of social media in the
democratic process.
|
2301.09060 | Ryan White | Basilio Caruso and Trupti Mahendrakar and Van Minh Nguyen and Ryan T.
White and Todd Steffen | 3D Reconstruction of Non-cooperative Resident Space Objects using
Instant NGP-accelerated NeRF and D-NeRF | Presented at AAS/AIAA Spaceflight Mechanics Conference 2023, 14
pages, 10 figures, 2 tables | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The proliferation of non-cooperative resident space objects (RSOs) in orbit
has spurred the demand for active space debris removal, on-orbit servicing
(OOS), classification, and functionality identification of these RSOs. Recent
advances in computer vision have enabled high-definition 3D modeling of objects
based on a set of 2D images captured from different viewing angles. This work
adapts Instant NeRF and D-NeRF, variations of the neural radiance field (NeRF)
algorithm to the problem of mapping RSOs in orbit for the purposes of
functionality identification and assisting with OOS. The algorithms are
evaluated for 3D reconstruction quality and hardware requirements using
datasets of images of a spacecraft mock-up taken under two different lighting
and motion conditions at the Orbital Robotic Interaction, On-Orbit Servicing
and Navigation (ORION) Laboratory at Florida Institute of Technology. Instant
NeRF is shown to learn high-fidelity 3D models with a computational cost that
could feasibly be trained on on-board computers.
| [
{
"version": "v1",
"created": "Sun, 22 Jan 2023 05:26:08 GMT"
},
{
"version": "v2",
"created": "Thu, 2 Feb 2023 02:31:40 GMT"
},
{
"version": "v3",
"created": "Fri, 9 Jun 2023 18:26:58 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Caruso",
"Basilio",
""
],
[
"Mahendrakar",
"Trupti",
""
],
[
"Nguyen",
"Van Minh",
""
],
[
"White",
"Ryan T.",
""
],
[
"Steffen",
"Todd",
""
]
] | TITLE: 3D Reconstruction of Non-cooperative Resident Space Objects using
Instant NGP-accelerated NeRF and D-NeRF
ABSTRACT: The proliferation of non-cooperative resident space objects (RSOs) in orbit
has spurred the demand for active space debris removal, on-orbit servicing
(OOS), classification, and functionality identification of these RSOs. Recent
advances in computer vision have enabled high-definition 3D modeling of objects
based on a set of 2D images captured from different viewing angles. This work
adapts Instant NeRF and D-NeRF, variations of the neural radiance field (NeRF)
algorithm to the problem of mapping RSOs in orbit for the purposes of
functionality identification and assisting with OOS. The algorithms are
evaluated for 3D reconstruction quality and hardware requirements using
datasets of images of a spacecraft mock-up taken under two different lighting
and motion conditions at the Orbital Robotic Interaction, On-Orbit Servicing
and Navigation (ORION) Laboratory at Florida Institute of Technology. Instant
NeRF is shown to learn high-fidelity 3D models with a computational cost that
could feasibly be trained on on-board computers.
|
2301.10900 | Xiaohu Huang | Xiaohu Huang, Hao Zhou, Jian Wang, Haocheng Feng, Junyu Han, Errui
Ding, Jingdong Wang, Xinggang Wang, Wenyu Liu, Bin Feng | Graph Contrastive Learning for Skeleton-based Action Recognition | Accepted by ICLR2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the field of skeleton-based action recognition, current top-performing
graph convolutional networks (GCNs) exploit intra-sequence context to construct
adaptive graphs for feature aggregation. However, we argue that such context is
still \textit{local} since the rich cross-sequence relations have not been
explicitly investigated. In this paper, we propose a graph contrastive learning
framework for skeleton-based action recognition (\textit{SkeletonGCL}) to
explore the \textit{global} context across all sequences. In specific,
SkeletonGCL associates graph learning across sequences by enforcing graphs to
be class-discriminative, \emph{i.e.,} intra-class compact and inter-class
dispersed, which improves the GCN capacity to distinguish various action
patterns. Besides, two memory banks are designed to enrich cross-sequence
context from two complementary levels, \emph{i.e.,} instance and semantic
levels, enabling graph contrastive learning in multiple context scales.
Consequently, SkeletonGCL establishes a new training paradigm, and it can be
seamlessly incorporated into current GCNs. Without loss of generality, we
combine SkeletonGCL with three GCNs (2S-ACGN, CTR-GCN, and InfoGCN), and
achieve consistent improvements on NTU60, NTU120, and NW-UCLA benchmarks. The
source code will be available at
\url{https://github.com/OliverHxh/SkeletonGCL}.
| [
{
"version": "v1",
"created": "Thu, 26 Jan 2023 02:09:16 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 10:32:06 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Huang",
"Xiaohu",
""
],
[
"Zhou",
"Hao",
""
],
[
"Wang",
"Jian",
""
],
[
"Feng",
"Haocheng",
""
],
[
"Han",
"Junyu",
""
],
[
"Ding",
"Errui",
""
],
[
"Wang",
"Jingdong",
""
],
[
"Wang",
"Xinggang",
""
],
[
"Liu",
"Wenyu",
""
],
[
"Feng",
"Bin",
""
]
] | TITLE: Graph Contrastive Learning for Skeleton-based Action Recognition
ABSTRACT: In the field of skeleton-based action recognition, current top-performing
graph convolutional networks (GCNs) exploit intra-sequence context to construct
adaptive graphs for feature aggregation. However, we argue that such context is
still \textit{local} since the rich cross-sequence relations have not been
explicitly investigated. In this paper, we propose a graph contrastive learning
framework for skeleton-based action recognition (\textit{SkeletonGCL}) to
explore the \textit{global} context across all sequences. In specific,
SkeletonGCL associates graph learning across sequences by enforcing graphs to
be class-discriminative, \emph{i.e.,} intra-class compact and inter-class
dispersed, which improves the GCN capacity to distinguish various action
patterns. Besides, two memory banks are designed to enrich cross-sequence
context from two complementary levels, \emph{i.e.,} instance and semantic
levels, enabling graph contrastive learning in multiple context scales.
Consequently, SkeletonGCL establishes a new training paradigm, and it can be
seamlessly incorporated into current GCNs. Without loss of generality, we
combine SkeletonGCL with three GCNs (2S-ACGN, CTR-GCN, and InfoGCN), and
achieve consistent improvements on NTU60, NTU120, and NW-UCLA benchmarks. The
source code will be available at
\url{https://github.com/OliverHxh/SkeletonGCL}.
|
2301.12972 | Sunghwan Yoo | Sunghwan Yoo, Yeongjeong Jeong, Maryam Jameela, Gunho Sohn | Human Vision Based 3D Point Cloud Semantic Segmentation of Large-Scale
Outdoor Scene | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | This paper proposes EyeNet, a novel semantic segmentation network for point
clouds that addresses the critical yet often overlooked parameter of coverage
area size. Inspired by human peripheral vision, EyeNet overcomes the
limitations of conventional networks by introducing a simple but efficient
multi-contour input and a parallel processing network with connection blocks
between parallel streams. The proposed approach effectively addresses the
challenges of dense point clouds, as demonstrated by our ablation studies and
state-of-the-art performance on Large-Scale Outdoor datasets.
| [
{
"version": "v1",
"created": "Mon, 30 Jan 2023 15:20:13 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Mar 2023 18:22:43 GMT"
},
{
"version": "v3",
"created": "Mon, 12 Jun 2023 16:54:05 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Yoo",
"Sunghwan",
""
],
[
"Jeong",
"Yeongjeong",
""
],
[
"Jameela",
"Maryam",
""
],
[
"Sohn",
"Gunho",
""
]
] | TITLE: Human Vision Based 3D Point Cloud Semantic Segmentation of Large-Scale
Outdoor Scene
ABSTRACT: This paper proposes EyeNet, a novel semantic segmentation network for point
clouds that addresses the critical yet often overlooked parameter of coverage
area size. Inspired by human peripheral vision, EyeNet overcomes the
limitations of conventional networks by introducing a simple but efficient
multi-contour input and a parallel processing network with connection blocks
between parallel streams. The proposed approach effectively addresses the
challenges of dense point clouds, as demonstrated by our ablation studies and
state-of-the-art performance on Large-Scale Outdoor datasets.
|
2302.00573 | Kersten Diers | Kersten Diers and Hannah Baumeister and Frank Jessen and Emrah D\"uzel
and David Berron and Martin Reuter | An automated, geometry-based method for hippocampal shape and thickness
analysis | Updated to journal publication | Neuroimage 276 (2023) 120182 | 10.1016/j.neuroimage.2023.120182 | null | cs.CV cs.CG | http://creativecommons.org/licenses/by/4.0/ | The hippocampus is one of the most studied neuroanatomical structures due to
its involvement in attention, learning, and memory as well as its atrophy in
ageing, neurological, and psychiatric diseases. Hippocampal shape changes,
however, are complex and cannot be fully characterized by a single summary
metric such as hippocampal volume as determined from MR images. In this work,
we propose an automated, geometry-based approach for the unfolding, point-wise
correspondence, and local analysis of hippocampal shape features such as
thickness and curvature. Starting from an automated segmentation of hippocampal
subfields, we create a 3D tetrahedral mesh model as well as a 3D intrinsic
coordinate system of the hippocampal body. From this coordinate system, we
derive local curvature and thickness estimates as well as a 2D sheet for
hippocampal unfolding. We evaluate the performance of our algorithm with a
series of experiments to quantify neurodegenerative changes in Mild Cognitive
Impairment and Alzheimer's disease dementia. We find that hippocampal thickness
estimates detect known differences between clinical groups and can determine
the location of these effects on the hippocampal sheet. Further, thickness
estimates improve classification of clinical groups and cognitively unimpaired
controls when added as an additional predictor. Comparable results are obtained
with different datasets and segmentation algorithms. Taken together, we
replicate canonical findings on hippocampal volume/shape changes in dementia,
extend them by gaining insight into their spatial localization on the
hippocampal sheet, and provide additional, complementary information beyond
traditional measures. We provide a new set of sensitive processing and analysis
tools for the analysis of hippocampal geometry that allows comparisons across
studies without relying on image registration or requiring manual intervention.
| [
{
"version": "v1",
"created": "Wed, 1 Feb 2023 16:46:46 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 12:21:34 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Diers",
"Kersten",
""
],
[
"Baumeister",
"Hannah",
""
],
[
"Jessen",
"Frank",
""
],
[
"Düzel",
"Emrah",
""
],
[
"Berron",
"David",
""
],
[
"Reuter",
"Martin",
""
]
] | TITLE: An automated, geometry-based method for hippocampal shape and thickness
analysis
ABSTRACT: The hippocampus is one of the most studied neuroanatomical structures due to
its involvement in attention, learning, and memory as well as its atrophy in
ageing, neurological, and psychiatric diseases. Hippocampal shape changes,
however, are complex and cannot be fully characterized by a single summary
metric such as hippocampal volume as determined from MR images. In this work,
we propose an automated, geometry-based approach for the unfolding, point-wise
correspondence, and local analysis of hippocampal shape features such as
thickness and curvature. Starting from an automated segmentation of hippocampal
subfields, we create a 3D tetrahedral mesh model as well as a 3D intrinsic
coordinate system of the hippocampal body. From this coordinate system, we
derive local curvature and thickness estimates as well as a 2D sheet for
hippocampal unfolding. We evaluate the performance of our algorithm with a
series of experiments to quantify neurodegenerative changes in Mild Cognitive
Impairment and Alzheimer's disease dementia. We find that hippocampal thickness
estimates detect known differences between clinical groups and can determine
the location of these effects on the hippocampal sheet. Further, thickness
estimates improve classification of clinical groups and cognitively unimpaired
controls when added as an additional predictor. Comparable results are obtained
with different datasets and segmentation algorithms. Taken together, we
replicate canonical findings on hippocampal volume/shape changes in dementia,
extend them by gaining insight into their spatial localization on the
hippocampal sheet, and provide additional, complementary information beyond
traditional measures. We provide a new set of sensitive processing and analysis
tools for the analysis of hippocampal geometry that allows comparisons across
studies without relying on image registration or requiring manual intervention.
|
2302.07612 | Ben Zandonati | Ben Zandonati, Glenn Bucagu, Adrian Alan Pol, Maurizio Pierini, Olya
Sirkin, Tal Kopetz | Towards Optimal Compression: Joint Pruning and Quantization | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Model compression is instrumental in optimizing deep neural network inference
on resource-constrained hardware. The prevailing methods for network
compression, namely quantization and pruning, have been shown to enhance
efficiency at the cost of performance. Determining the most effective
quantization and pruning strategies for individual layers and parameters
remains a challenging problem, often requiring computationally expensive and ad
hoc numerical optimization techniques. This paper introduces FITCompress, a
novel method integrating layer-wise mixed-precision quantization and
unstructured pruning using a unified heuristic approach. By leveraging the
Fisher Information Metric and path planning through compression space,
FITCompress optimally selects a combination of pruning mask and mixed-precision
quantization configuration for a given pre-trained model and compression
constraint. Experiments on computer vision and natural language processing
benchmarks demonstrate that our proposed approach achieves a superior
compression-performance trade-off compared to existing state-of-the-art
methods. FITCompress stands out for its principled derivation, making it
versatile across tasks and network architectures, and represents a step towards
achieving optimal compression for neural networks.
| [
{
"version": "v1",
"created": "Wed, 15 Feb 2023 12:02:30 GMT"
},
{
"version": "v2",
"created": "Sun, 11 Jun 2023 10:01:07 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zandonati",
"Ben",
""
],
[
"Bucagu",
"Glenn",
""
],
[
"Pol",
"Adrian Alan",
""
],
[
"Pierini",
"Maurizio",
""
],
[
"Sirkin",
"Olya",
""
],
[
"Kopetz",
"Tal",
""
]
] | TITLE: Towards Optimal Compression: Joint Pruning and Quantization
ABSTRACT: Model compression is instrumental in optimizing deep neural network inference
on resource-constrained hardware. The prevailing methods for network
compression, namely quantization and pruning, have been shown to enhance
efficiency at the cost of performance. Determining the most effective
quantization and pruning strategies for individual layers and parameters
remains a challenging problem, often requiring computationally expensive and ad
hoc numerical optimization techniques. This paper introduces FITCompress, a
novel method integrating layer-wise mixed-precision quantization and
unstructured pruning using a unified heuristic approach. By leveraging the
Fisher Information Metric and path planning through compression space,
FITCompress optimally selects a combination of pruning mask and mixed-precision
quantization configuration for a given pre-trained model and compression
constraint. Experiments on computer vision and natural language processing
benchmarks demonstrate that our proposed approach achieves a superior
compression-performance trade-off compared to existing state-of-the-art
methods. FITCompress stands out for its principled derivation, making it
versatile across tasks and network architectures, and represents a step towards
achieving optimal compression for neural networks.
|
2302.08237 | Ahmed Alia | Ahmed Alia, Mohammed Maree, Mohcine Chraibi, Anas Toma and Armin
Seyfried | A Cloud-based Deep Learning Framework for Early Detection of Pushing at
Crowded Event Entrances | null | 2023, IEEE Access | 10.1109/ACCESS.2023.3273770 | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Crowding at the entrances of large events may lead to critical and
life-threatening situations, particularly when people start pushing each other
to reach the event faster. Automatic and timely identification of pushing
behavior would help organizers and security forces to intervene early and
mitigate dangerous situations. In this paper, we propose a cloud-based deep
learning framework for automatic early detection of pushing in crowded event
entrances. The proposed framework initially modifies and trains the
EfficientNetV2B0 Convolutional Neural Network model. Subsequently, it
integrates the adapted model with an accurate and fast pre-trained deep optical
flow model with the color wheel method to analyze video streams and identify
pushing patches in real-time. Moreover, the framework uses live capturing
technology and a cloud-based environment to collect video streams of crowds in
real-time and provide early-stage results. A novel dataset is generated based
on five real-world experiments and their associated ground truth data to train
the adapted EfficientNetV2B0 model. The experimental setups simulated a crowded
event entrance, while the ground truths for each video experiment was generated
manually by social psychologists. Several experiments on the videos and the
generated dataset are carried out to evaluate the accuracy and annotation delay
time of the proposed framework. The experimental results show that the proposed
framework identified pushing behaviors with an accuracy rate of 87% within a
reasonable delay time.
| [
{
"version": "v1",
"created": "Thu, 16 Feb 2023 11:39:32 GMT"
},
{
"version": "v2",
"created": "Fri, 9 Jun 2023 19:45:42 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Alia",
"Ahmed",
""
],
[
"Maree",
"Mohammed",
""
],
[
"Chraibi",
"Mohcine",
""
],
[
"Toma",
"Anas",
""
],
[
"Seyfried",
"Armin",
""
]
] | TITLE: A Cloud-based Deep Learning Framework for Early Detection of Pushing at
Crowded Event Entrances
ABSTRACT: Crowding at the entrances of large events may lead to critical and
life-threatening situations, particularly when people start pushing each other
to reach the event faster. Automatic and timely identification of pushing
behavior would help organizers and security forces to intervene early and
mitigate dangerous situations. In this paper, we propose a cloud-based deep
learning framework for automatic early detection of pushing in crowded event
entrances. The proposed framework initially modifies and trains the
EfficientNetV2B0 Convolutional Neural Network model. Subsequently, it
integrates the adapted model with an accurate and fast pre-trained deep optical
flow model with the color wheel method to analyze video streams and identify
pushing patches in real-time. Moreover, the framework uses live capturing
technology and a cloud-based environment to collect video streams of crowds in
real-time and provide early-stage results. A novel dataset is generated based
on five real-world experiments and their associated ground truth data to train
the adapted EfficientNetV2B0 model. The experimental setups simulated a crowded
event entrance, while the ground truths for each video experiment was generated
manually by social psychologists. Several experiments on the videos and the
generated dataset are carried out to evaluate the accuracy and annotation delay
time of the proposed framework. The experimental results show that the proposed
framework identified pushing behaviors with an accuracy rate of 87% within a
reasonable delay time.
|
2302.10911 | Zexi Li | Zexi Li, Tao Lin, Xinyi Shang, Chao Wu | Revisiting Weighted Aggregation in Federated Learning with Neural
Networks | Accepted by ICML 2023 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In federated learning (FL), weighted aggregation of local models is conducted
to generate a global model, and the aggregation weights are normalized (the sum
of weights is 1) and proportional to the local data sizes. In this paper, we
revisit the weighted aggregation process and gain new insights into the
training dynamics of FL. First, we find that the sum of weights can be smaller
than 1, causing global weight shrinking effect (analogous to weight decay) and
improving generalization. We explore how the optimal shrinking factor is
affected by clients' data heterogeneity and local epochs. Second, we dive into
the relative aggregation weights among clients to depict the clients'
importance. We develop client coherence to study the learning dynamics and find
a critical point that exists. Before entering the critical point, more coherent
clients play more essential roles in generalization. Based on the above
insights, we propose an effective method for Federated Learning with Learnable
Aggregation Weights, named as FedLAW. Extensive experiments verify that our
method can improve the generalization of the global model by a large margin on
different datasets and models.
| [
{
"version": "v1",
"created": "Tue, 14 Feb 2023 07:52:21 GMT"
},
{
"version": "v2",
"created": "Tue, 16 May 2023 13:03:14 GMT"
},
{
"version": "v3",
"created": "Fri, 9 Jun 2023 14:49:41 GMT"
},
{
"version": "v4",
"created": "Mon, 12 Jun 2023 14:19:53 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Li",
"Zexi",
""
],
[
"Lin",
"Tao",
""
],
[
"Shang",
"Xinyi",
""
],
[
"Wu",
"Chao",
""
]
] | TITLE: Revisiting Weighted Aggregation in Federated Learning with Neural
Networks
ABSTRACT: In federated learning (FL), weighted aggregation of local models is conducted
to generate a global model, and the aggregation weights are normalized (the sum
of weights is 1) and proportional to the local data sizes. In this paper, we
revisit the weighted aggregation process and gain new insights into the
training dynamics of FL. First, we find that the sum of weights can be smaller
than 1, causing global weight shrinking effect (analogous to weight decay) and
improving generalization. We explore how the optimal shrinking factor is
affected by clients' data heterogeneity and local epochs. Second, we dive into
the relative aggregation weights among clients to depict the clients'
importance. We develop client coherence to study the learning dynamics and find
a critical point that exists. Before entering the critical point, more coherent
clients play more essential roles in generalization. Based on the above
insights, we propose an effective method for Federated Learning with Learnable
Aggregation Weights, named as FedLAW. Extensive experiments verify that our
method can improve the generalization of the global model by a large margin on
different datasets and models.
|
2302.13726 | Haoyi Niu | Haoyi Niu, Kun Ren, Yizhou Xu, Ziyuan Yang, Yichen Lin, Yi Zhang,
Jianming Hu | (Re)$^2$H2O: Autonomous Driving Scenario Generation via Reversely
Regularized Hybrid Offline-and-Online Reinforcement Learning | Accepted in IEEE Intelligent Vehicles Symposium 2023 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Autonomous driving and its widespread adoption have long held tremendous
promise. Nevertheless, without a trustworthy and thorough testing procedure,
not only does the industry struggle to mass-produce autonomous vehicles (AV),
but neither the general public nor policymakers are convinced to accept the
innovations. Generating safety-critical scenarios that present significant
challenges to AV is an essential first step in testing. Real-world datasets
include naturalistic but overly safe driving behaviors, whereas simulation
would allow for unrestricted exploration of diverse and aggressive traffic
scenarios. Conversely, higher-dimensional searching space in simulation
disables efficient scenario generation without real-world data distribution as
implicit constraints. In order to marry the benefits of both, it seems
appealing to learn to generate scenarios from both offline real-world and
online simulation data simultaneously. Therefore, we tailor a Reversely
Regularized Hybrid Offline-and-Online ((Re)$^2$H2O) Reinforcement Learning
recipe to additionally penalize Q-values on real-world data and reward Q-values
on simulated data, which ensures the generated scenarios are both varied and
adversarial. Through extensive experiments, our solution proves to produce more
risky scenarios than competitive baselines and it can generalize to work with
various autonomous driving models. In addition, these generated scenarios are
also corroborated to be capable of fine-tuning AV performance.
| [
{
"version": "v1",
"created": "Mon, 27 Feb 2023 12:44:48 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 09:27:51 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Niu",
"Haoyi",
""
],
[
"Ren",
"Kun",
""
],
[
"Xu",
"Yizhou",
""
],
[
"Yang",
"Ziyuan",
""
],
[
"Lin",
"Yichen",
""
],
[
"Zhang",
"Yi",
""
],
[
"Hu",
"Jianming",
""
]
] | TITLE: (Re)$^2$H2O: Autonomous Driving Scenario Generation via Reversely
Regularized Hybrid Offline-and-Online Reinforcement Learning
ABSTRACT: Autonomous driving and its widespread adoption have long held tremendous
promise. Nevertheless, without a trustworthy and thorough testing procedure,
not only does the industry struggle to mass-produce autonomous vehicles (AV),
but neither the general public nor policymakers are convinced to accept the
innovations. Generating safety-critical scenarios that present significant
challenges to AV is an essential first step in testing. Real-world datasets
include naturalistic but overly safe driving behaviors, whereas simulation
would allow for unrestricted exploration of diverse and aggressive traffic
scenarios. Conversely, higher-dimensional searching space in simulation
disables efficient scenario generation without real-world data distribution as
implicit constraints. In order to marry the benefits of both, it seems
appealing to learn to generate scenarios from both offline real-world and
online simulation data simultaneously. Therefore, we tailor a Reversely
Regularized Hybrid Offline-and-Online ((Re)$^2$H2O) Reinforcement Learning
recipe to additionally penalize Q-values on real-world data and reward Q-values
on simulated data, which ensures the generated scenarios are both varied and
adversarial. Through extensive experiments, our solution proves to produce more
risky scenarios than competitive baselines and it can generalize to work with
various autonomous driving models. In addition, these generated scenarios are
also corroborated to be capable of fine-tuning AV performance.
|
2302.14555 | Yannick Wack | Yannick Wack, Sylvain Serra, Martine Baelmans, Jean-Michel Reneaume,
Maarten Blommaert | Non-linear Topology Optimization of District Heating Networks: A
benchmark of Mixed-Integer and Adjoint Approaches | null | Energy Volume 278, Part B, 1 September 2023, 127977 | 10.1016/j.energy.2023.127977 | null | math.OC cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | The widespread use of optimization methods in the design phase of District
Heating Networks is currently limited by the availability of scalable
optimization approaches that accurately represent the network. In this paper,
we compare and benchmark two different approaches to non-linear topology
optimization of District Heating Networks in terms of computational cost and
optimality gap. The first approach solves a mixed-integer non-linear
optimization problem that resolves the binary constraints of pipe routing
choices using a combinatorial optimization approach. The second approach solves
a relaxed optimization problem using an adjoint optimization approach, and
enforces a discrete network topology through penalization. Our benchmark shows
that the relaxed penalized problem has a polynomial computational cost scaling,
while the combinatorial solution scales exponentially, making it intractable
for practical-sized networks. We also evaluate the optimality gap between the
two approaches on two different District Heating Network optimization cases. We
find that the mixed-integer approach outperforms the adjoint approach on a
single-producer case, but the relaxed penalized problem is superior on a
multi-producer case. Based on this study, we discuss the importance of
initialization strategies for solving the optimal topology and design problem
of District Heating Networks as a non-linear optimization problem.
| [
{
"version": "v1",
"created": "Tue, 28 Feb 2023 13:19:16 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Wack",
"Yannick",
""
],
[
"Serra",
"Sylvain",
""
],
[
"Baelmans",
"Martine",
""
],
[
"Reneaume",
"Jean-Michel",
""
],
[
"Blommaert",
"Maarten",
""
]
] | TITLE: Non-linear Topology Optimization of District Heating Networks: A
benchmark of Mixed-Integer and Adjoint Approaches
ABSTRACT: The widespread use of optimization methods in the design phase of District
Heating Networks is currently limited by the availability of scalable
optimization approaches that accurately represent the network. In this paper,
we compare and benchmark two different approaches to non-linear topology
optimization of District Heating Networks in terms of computational cost and
optimality gap. The first approach solves a mixed-integer non-linear
optimization problem that resolves the binary constraints of pipe routing
choices using a combinatorial optimization approach. The second approach solves
a relaxed optimization problem using an adjoint optimization approach, and
enforces a discrete network topology through penalization. Our benchmark shows
that the relaxed penalized problem has a polynomial computational cost scaling,
while the combinatorial solution scales exponentially, making it intractable
for practical-sized networks. We also evaluate the optimality gap between the
two approaches on two different District Heating Network optimization cases. We
find that the mixed-integer approach outperforms the adjoint approach on a
single-producer case, but the relaxed penalized problem is superior on a
multi-producer case. Based on this study, we discuss the importance of
initialization strategies for solving the optimal topology and design problem
of District Heating Networks as a non-linear optimization problem.
|
2302.14685 | Samyak Jain | Samyak Jain, Sravanti Addepalli, Pawan Sahu, Priyam Dey and R.
Venkatesh Babu | DART: Diversify-Aggregate-Repeat Training Improves Generalization of
Neural Networks | CVPR 2023. First two authors contributed equally | null | null | null | cs.LG cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generalization of neural networks is crucial for deploying them safely in the
real world. Common training strategies to improve generalization involve the
use of data augmentations, ensembling and model averaging. In this work, we
first establish a surprisingly simple but strong benchmark for generalization
which utilizes diverse augmentations within a training minibatch, and show that
this can learn a more balanced distribution of features. Further, we propose
Diversify-Aggregate-Repeat Training (DART) strategy that first trains diverse
models using different augmentations (or domains) to explore the loss basin,
and further Aggregates their weights to combine their expertise and obtain
improved generalization. We find that Repeating the step of Aggregation
throughout training improves the overall optimization trajectory and also
ensures that the individual models have a sufficiently low loss barrier to
obtain improved generalization on combining them. We shed light on our approach
by casting it in the framework proposed by Shen et al. and theoretically show
that it indeed generalizes better. In addition to improvements in In- Domain
generalization, we demonstrate SOTA performance on the Domain Generalization
benchmarks in the popular DomainBed framework as well. Our method is generic
and can easily be integrated with several base training algorithms to achieve
performance gains.
| [
{
"version": "v1",
"created": "Tue, 28 Feb 2023 15:54:47 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 15:11:02 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Jain",
"Samyak",
""
],
[
"Addepalli",
"Sravanti",
""
],
[
"Sahu",
"Pawan",
""
],
[
"Dey",
"Priyam",
""
],
[
"Babu",
"R. Venkatesh",
""
]
] | TITLE: DART: Diversify-Aggregate-Repeat Training Improves Generalization of
Neural Networks
ABSTRACT: Generalization of neural networks is crucial for deploying them safely in the
real world. Common training strategies to improve generalization involve the
use of data augmentations, ensembling and model averaging. In this work, we
first establish a surprisingly simple but strong benchmark for generalization
which utilizes diverse augmentations within a training minibatch, and show that
this can learn a more balanced distribution of features. Further, we propose
Diversify-Aggregate-Repeat Training (DART) strategy that first trains diverse
models using different augmentations (or domains) to explore the loss basin,
and further Aggregates their weights to combine their expertise and obtain
improved generalization. We find that Repeating the step of Aggregation
throughout training improves the overall optimization trajectory and also
ensures that the individual models have a sufficiently low loss barrier to
obtain improved generalization on combining them. We shed light on our approach
by casting it in the framework proposed by Shen et al. and theoretically show
that it indeed generalizes better. In addition to improvements in In- Domain
generalization, we demonstrate SOTA performance on the Domain Generalization
benchmarks in the popular DomainBed framework as well. Our method is generic
and can easily be integrated with several base training algorithms to achieve
performance gains.
|
2303.05203 | Xiuyu Yang | Xiuyu Yang, Zhuangyan Zhang, Haikuo Du, Sui Yang, Fengping Sun, Yanbo
Liu, Ling Pei, Wenchao Xu, Weiqi Sun, Zhengyu Li | RMMDet: Road-Side Multitype and Multigroup Sensor Detection System for
Autonomous Driving | null | null | null | null | cs.RO cs.AI cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Autonomous driving has now made great strides thanks to artificial
intelligence, and numerous advanced methods have been proposed for vehicle end
target detection, including single sensor or multi sensor detection methods.
However, the complexity and diversity of real traffic situations necessitate an
examination of how to use these methods in real road conditions. In this paper,
we propose RMMDet, a road-side multitype and multigroup sensor detection system
for autonomous driving. We use a ROS-based virtual environment to simulate
real-world conditions, in particular the physical and functional construction
of the sensors. Then we implement muti-type sensor detection and multi-group
sensors fusion in this environment, including camera-radar and camera-lidar
detection based on result-level fusion. We produce local datasets and real sand
table field, and conduct various experiments. Furthermore, we link a
multi-agent collaborative scheduling system to the fusion detection system.
Hence, the whole roadside detection system is formed by roadside perception,
fusion detection, and scheduling planning. Through the experiments, it can be
seen that RMMDet system we built plays an important role in vehicle-road
collaboration and its optimization. The code and supplementary materials can be
found at: https://github.com/OrangeSodahub/RMMDet
| [
{
"version": "v1",
"created": "Thu, 9 Mar 2023 12:13:39 GMT"
},
{
"version": "v2",
"created": "Fri, 10 Mar 2023 03:19:20 GMT"
},
{
"version": "v3",
"created": "Sat, 10 Jun 2023 01:07:03 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Yang",
"Xiuyu",
""
],
[
"Zhang",
"Zhuangyan",
""
],
[
"Du",
"Haikuo",
""
],
[
"Yang",
"Sui",
""
],
[
"Sun",
"Fengping",
""
],
[
"Liu",
"Yanbo",
""
],
[
"Pei",
"Ling",
""
],
[
"Xu",
"Wenchao",
""
],
[
"Sun",
"Weiqi",
""
],
[
"Li",
"Zhengyu",
""
]
] | TITLE: RMMDet: Road-Side Multitype and Multigroup Sensor Detection System for
Autonomous Driving
ABSTRACT: Autonomous driving has now made great strides thanks to artificial
intelligence, and numerous advanced methods have been proposed for vehicle end
target detection, including single sensor or multi sensor detection methods.
However, the complexity and diversity of real traffic situations necessitate an
examination of how to use these methods in real road conditions. In this paper,
we propose RMMDet, a road-side multitype and multigroup sensor detection system
for autonomous driving. We use a ROS-based virtual environment to simulate
real-world conditions, in particular the physical and functional construction
of the sensors. Then we implement muti-type sensor detection and multi-group
sensors fusion in this environment, including camera-radar and camera-lidar
detection based on result-level fusion. We produce local datasets and real sand
table field, and conduct various experiments. Furthermore, we link a
multi-agent collaborative scheduling system to the fusion detection system.
Hence, the whole roadside detection system is formed by roadside perception,
fusion detection, and scheduling planning. Through the experiments, it can be
seen that RMMDet system we built plays an important role in vehicle-road
collaboration and its optimization. The code and supplementary materials can be
found at: https://github.com/OrangeSodahub/RMMDet
|
2303.05470 | Gbetondji Dovonon | Aengus Lynch, Gb\`etondji J-S Dovonon, Jean Kaddour, Ricardo Silva | Spawrious: A Benchmark for Fine Control of Spurious Correlation Biases | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | The problem of spurious correlations (SCs) arises when a classifier relies on
non-predictive features that happen to be correlated with the labels in the
training data. For example, a classifier may misclassify dog breeds based on
the background of dog images. This happens when the backgrounds are correlated
with other breeds in the training data, leading to misclassifications during
test time. Previous SC benchmark datasets suffer from varying issues, e.g.,
over-saturation or only containing one-to-one (O2O) SCs, but no many-to-many
(M2M) SCs arising between groups of spurious attributes and classes. In this
paper, we present \benchmark-\{O2O, M2M\}-\{Easy, Medium, Hard\}, an image
classification benchmark suite containing spurious correlations between classes
and backgrounds. To create this dataset, we employ a text-to-image model to
generate photo-realistic images and an image captioning model to filter out
unsuitable ones. The resulting dataset is of high quality and contains
approximately 152k images. Our experimental results demonstrate that
state-of-the-art group robustness methods struggle with \benchmark, most
notably on the Hard-splits with none of them getting over $70\%$ accuracy on
the hardest split using a ResNet50 pretrained on ImageNet. By examining model
misclassifications, we detect reliances on spurious backgrounds, demonstrating
that our dataset provides a significant challenge.
| [
{
"version": "v1",
"created": "Thu, 9 Mar 2023 18:22:12 GMT"
},
{
"version": "v2",
"created": "Thu, 8 Jun 2023 18:00:46 GMT"
},
{
"version": "v3",
"created": "Mon, 12 Jun 2023 14:04:53 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Lynch",
"Aengus",
""
],
[
"Dovonon",
"Gbètondji J-S",
""
],
[
"Kaddour",
"Jean",
""
],
[
"Silva",
"Ricardo",
""
]
] | TITLE: Spawrious: A Benchmark for Fine Control of Spurious Correlation Biases
ABSTRACT: The problem of spurious correlations (SCs) arises when a classifier relies on
non-predictive features that happen to be correlated with the labels in the
training data. For example, a classifier may misclassify dog breeds based on
the background of dog images. This happens when the backgrounds are correlated
with other breeds in the training data, leading to misclassifications during
test time. Previous SC benchmark datasets suffer from varying issues, e.g.,
over-saturation or only containing one-to-one (O2O) SCs, but no many-to-many
(M2M) SCs arising between groups of spurious attributes and classes. In this
paper, we present \benchmark-\{O2O, M2M\}-\{Easy, Medium, Hard\}, an image
classification benchmark suite containing spurious correlations between classes
and backgrounds. To create this dataset, we employ a text-to-image model to
generate photo-realistic images and an image captioning model to filter out
unsuitable ones. The resulting dataset is of high quality and contains
approximately 152k images. Our experimental results demonstrate that
state-of-the-art group robustness methods struggle with \benchmark, most
notably on the Hard-splits with none of them getting over $70\%$ accuracy on
the hardest split using a ResNet50 pretrained on ImageNet. By examining model
misclassifications, we detect reliances on spurious backgrounds, demonstrating
that our dataset provides a significant challenge.
|
2303.06311 | Igor Ostrovskiy | S. Li, I. Ostrovskiy, Z. Li, L. Yang, S. Al Kharusi, G. Anton, I.
Badhrees, P.S. Barbeau, D. Beck, V. Belov, T. Bhatta, M. Breidenbach, T.
Brunner, G.F. Cao, W.R. Cen, C. Chambers, B. Cleveland, M. Coon, A.
Craycraft, T. Daniels, L. Darroch, S.J. Daugherty, J. Davis, S. Delaquis, A.
Der Mesrobian-Kabakian, R. DeVoe, J. Dilling, A. Dolgolenko, M.J. Dolinski,
J. Echevers, W. Fairbank Jr., D. Fairbank, J. Farine, S. Feyzbakhsh, P.
Fierlinger, Y.S. Fu, D. Fudenberg, P. Gautam, R. Gornea, G. Gratta, C. Hall,
E.V. Hansen, J. Hoessl, P. Hufschmidt, M. Hughes, A. Iverson, A. Jamil, C.
Jessiman, M.J. Jewell, A. Johnson, A. Karelin, L.J. Kaufman, T. Koffas, R.
Kr\"ucken, A. Kuchenkov, K.S. Kumar, Y. Lan, A. Larson, B.G. Lenardo, D.S.
Leonard, G.S. Li, C. Licciardi, Y.H. Lin, R. MacLellan, T. McElroy, T.
Michel, B. Mong, D.C. Moore, K. Murray, O. Njoya, O. Nusair, A. Odian, A.
Perna, A. Piepke, A. Pocar, F. Reti\`ere, A.L. Robinson, P.C. Rowson, J.
Runge, S. Schmidt, D. Sinclair, K. Skarpaas, A.K. Soma, V. Stekhanov, M.
Tarka, S. Thibado, J. Todd, T. Tolba, T.I. Totev, R. Tsang | Generative Adversarial Networks for Scintillation Signal Simulation in
EXO-200 | As accepted by JINST | JINST 18 P06005 2023 | 10.1088/1748-0221/18/06/P06005 | null | hep-ex cs.LG physics.ins-det | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generative Adversarial Networks trained on samples of simulated or actual
events have been proposed as a way of generating large simulated datasets at a
reduced computational cost. In this work, a novel approach to perform the
simulation of photodetector signals from the time projection chamber of the
EXO-200 experiment is demonstrated. The method is based on a Wasserstein
Generative Adversarial Network - a deep learning technique allowing for
implicit non-parametric estimation of the population distribution for a given
set of objects. Our network is trained on real calibration data using raw
scintillation waveforms as input. We find that it is able to produce
high-quality simulated waveforms an order of magnitude faster than the
traditional simulation approach and, importantly, generalize from the training
sample and discern salient high-level features of the data. In particular, the
network correctly deduces position dependency of scintillation light response
in the detector and correctly recognizes dead photodetector channels. The
network output is then integrated into the EXO-200 analysis framework to show
that the standard EXO-200 reconstruction routine processes the simulated
waveforms to produce energy distributions comparable to that of real waveforms.
Finally, the remaining discrepancies and potential ways to improve the approach
further are highlighted.
| [
{
"version": "v1",
"created": "Sat, 11 Mar 2023 05:16:20 GMT"
},
{
"version": "v2",
"created": "Mon, 8 May 2023 13:57:24 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Li",
"S.",
""
],
[
"Ostrovskiy",
"I.",
""
],
[
"Li",
"Z.",
""
],
[
"Yang",
"L.",
""
],
[
"Kharusi",
"S. Al",
""
],
[
"Anton",
"G.",
""
],
[
"Badhrees",
"I.",
""
],
[
"Barbeau",
"P. S.",
""
],
[
"Beck",
"D.",
""
],
[
"Belov",
"V.",
""
],
[
"Bhatta",
"T.",
""
],
[
"Breidenbach",
"M.",
""
],
[
"Brunner",
"T.",
""
],
[
"Cao",
"G. F.",
""
],
[
"Cen",
"W. R.",
""
],
[
"Chambers",
"C.",
""
],
[
"Cleveland",
"B.",
""
],
[
"Coon",
"M.",
""
],
[
"Craycraft",
"A.",
""
],
[
"Daniels",
"T.",
""
],
[
"Darroch",
"L.",
""
],
[
"Daugherty",
"S. J.",
""
],
[
"Davis",
"J.",
""
],
[
"Delaquis",
"S.",
""
],
[
"Der Mesrobian-Kabakian",
"A.",
""
],
[
"DeVoe",
"R.",
""
],
[
"Dilling",
"J.",
""
],
[
"Dolgolenko",
"A.",
""
],
[
"Dolinski",
"M. J.",
""
],
[
"Echevers",
"J.",
""
],
[
"Fairbank",
"W.",
"Jr."
],
[
"Fairbank",
"D.",
""
],
[
"Farine",
"J.",
""
],
[
"Feyzbakhsh",
"S.",
""
],
[
"Fierlinger",
"P.",
""
],
[
"Fu",
"Y. S.",
""
],
[
"Fudenberg",
"D.",
""
],
[
"Gautam",
"P.",
""
],
[
"Gornea",
"R.",
""
],
[
"Gratta",
"G.",
""
],
[
"Hall",
"C.",
""
],
[
"Hansen",
"E. V.",
""
],
[
"Hoessl",
"J.",
""
],
[
"Hufschmidt",
"P.",
""
],
[
"Hughes",
"M.",
""
],
[
"Iverson",
"A.",
""
],
[
"Jamil",
"A.",
""
],
[
"Jessiman",
"C.",
""
],
[
"Jewell",
"M. J.",
""
],
[
"Johnson",
"A.",
""
],
[
"Karelin",
"A.",
""
],
[
"Kaufman",
"L. J.",
""
],
[
"Koffas",
"T.",
""
],
[
"Krücken",
"R.",
""
],
[
"Kuchenkov",
"A.",
""
],
[
"Kumar",
"K. S.",
""
],
[
"Lan",
"Y.",
""
],
[
"Larson",
"A.",
""
],
[
"Lenardo",
"B. G.",
""
],
[
"Leonard",
"D. S.",
""
],
[
"Li",
"G. S.",
""
],
[
"Licciardi",
"C.",
""
],
[
"Lin",
"Y. H.",
""
],
[
"MacLellan",
"R.",
""
],
[
"McElroy",
"T.",
""
],
[
"Michel",
"T.",
""
],
[
"Mong",
"B.",
""
],
[
"Moore",
"D. C.",
""
],
[
"Murray",
"K.",
""
],
[
"Njoya",
"O.",
""
],
[
"Nusair",
"O.",
""
],
[
"Odian",
"A.",
""
],
[
"Perna",
"A.",
""
],
[
"Piepke",
"A.",
""
],
[
"Pocar",
"A.",
""
],
[
"Retière",
"F.",
""
],
[
"Robinson",
"A. L.",
""
],
[
"Rowson",
"P. C.",
""
],
[
"Runge",
"J.",
""
],
[
"Schmidt",
"S.",
""
],
[
"Sinclair",
"D.",
""
],
[
"Skarpaas",
"K.",
""
],
[
"Soma",
"A. K.",
""
],
[
"Stekhanov",
"V.",
""
],
[
"Tarka",
"M.",
""
],
[
"Thibado",
"S.",
""
],
[
"Todd",
"J.",
""
],
[
"Tolba",
"T.",
""
],
[
"Totev",
"T. I.",
""
],
[
"Tsang",
"R.",
""
]
] | TITLE: Generative Adversarial Networks for Scintillation Signal Simulation in
EXO-200
ABSTRACT: Generative Adversarial Networks trained on samples of simulated or actual
events have been proposed as a way of generating large simulated datasets at a
reduced computational cost. In this work, a novel approach to perform the
simulation of photodetector signals from the time projection chamber of the
EXO-200 experiment is demonstrated. The method is based on a Wasserstein
Generative Adversarial Network - a deep learning technique allowing for
implicit non-parametric estimation of the population distribution for a given
set of objects. Our network is trained on real calibration data using raw
scintillation waveforms as input. We find that it is able to produce
high-quality simulated waveforms an order of magnitude faster than the
traditional simulation approach and, importantly, generalize from the training
sample and discern salient high-level features of the data. In particular, the
network correctly deduces position dependency of scintillation light response
in the detector and correctly recognizes dead photodetector channels. The
network output is then integrated into the EXO-200 analysis framework to show
that the standard EXO-200 reconstruction routine processes the simulated
waveforms to produce energy distributions comparable to that of real waveforms.
Finally, the remaining discrepancies and potential ways to improve the approach
further are highlighted.
|
2303.06865 | Ying Sheng | Ying Sheng, Lianmin Zheng, Binhang Yuan, Zhuohan Li, Max Ryabinin,
Daniel Y. Fu, Zhiqiang Xie, Beidi Chen, Clark Barrett, Joseph E. Gonzalez,
Percy Liang, Christopher R\'e, Ion Stoica, Ce Zhang | FlexGen: High-Throughput Generative Inference of Large Language Models
with a Single GPU | null | null | null | null | cs.LG cs.AI cs.PF | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The high computational and memory requirements of large language model (LLM)
inference make it feasible only with multiple high-end accelerators. Motivated
by the emerging demand for latency-insensitive tasks with batched processing,
this paper initiates the study of high-throughput LLM inference using limited
resources, such as a single commodity GPU. We present FlexGen, a
high-throughput generation engine for running LLMs with limited GPU memory.
FlexGen can be flexibly configured under various hardware resource constraints
by aggregating memory and computation from the GPU, CPU, and disk. By solving a
linear programming problem, it searches for efficient patterns to store and
access tensors. FlexGen further compresses the weights and the attention cache
to 4 bits with negligible accuracy loss. These techniques enable FlexGen to
have a larger space of batch size choices and thus significantly increase
maximum throughput. As a result, when running OPT-175B on a single 16GB GPU,
FlexGen achieves significantly higher throughput compared to state-of-the-art
offloading systems, reaching a generation throughput of 1 token/s for the first
time with an effective batch size of 144. On the HELM benchmark, FlexGen can
benchmark a 30B model with a 16GB GPU on 7 representative sub-scenarios in 21
hours. The code is available at https://github.com/FMInference/FlexGen
| [
{
"version": "v1",
"created": "Mon, 13 Mar 2023 05:19:28 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 07:48:53 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Sheng",
"Ying",
""
],
[
"Zheng",
"Lianmin",
""
],
[
"Yuan",
"Binhang",
""
],
[
"Li",
"Zhuohan",
""
],
[
"Ryabinin",
"Max",
""
],
[
"Fu",
"Daniel Y.",
""
],
[
"Xie",
"Zhiqiang",
""
],
[
"Chen",
"Beidi",
""
],
[
"Barrett",
"Clark",
""
],
[
"Gonzalez",
"Joseph E.",
""
],
[
"Liang",
"Percy",
""
],
[
"Ré",
"Christopher",
""
],
[
"Stoica",
"Ion",
""
],
[
"Zhang",
"Ce",
""
]
] | TITLE: FlexGen: High-Throughput Generative Inference of Large Language Models
with a Single GPU
ABSTRACT: The high computational and memory requirements of large language model (LLM)
inference make it feasible only with multiple high-end accelerators. Motivated
by the emerging demand for latency-insensitive tasks with batched processing,
this paper initiates the study of high-throughput LLM inference using limited
resources, such as a single commodity GPU. We present FlexGen, a
high-throughput generation engine for running LLMs with limited GPU memory.
FlexGen can be flexibly configured under various hardware resource constraints
by aggregating memory and computation from the GPU, CPU, and disk. By solving a
linear programming problem, it searches for efficient patterns to store and
access tensors. FlexGen further compresses the weights and the attention cache
to 4 bits with negligible accuracy loss. These techniques enable FlexGen to
have a larger space of batch size choices and thus significantly increase
maximum throughput. As a result, when running OPT-175B on a single 16GB GPU,
FlexGen achieves significantly higher throughput compared to state-of-the-art
offloading systems, reaching a generation throughput of 1 token/s for the first
time with an effective batch size of 144. On the HELM benchmark, FlexGen can
benchmark a 30B model with a 16GB GPU on 7 representative sub-scenarios in 21
hours. The code is available at https://github.com/FMInference/FlexGen
|
2303.07667 | Ganghui Ru | Ganghui Ru, Xulong Zhang, Jianzong Wang, Ning Cheng, Jing Xiao | Improving Music Genre Classification from Multi-Modal Properties of
Music and Genre Correlations Perspective | Accepted by ICASSP 2023 | null | null | null | cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Music genre classification has been widely studied in past few years for its
various applications in music information retrieval. Previous works tend to
perform unsatisfactorily, since those methods only use audio content or jointly
use audio content and lyrics content inefficiently. In addition, as genres
normally co-occur in a music track, it is desirable to capture and model the
genre correlations to improve the performance of multi-label music genre
classification. To solve these issues, we present a novel multi-modal method
leveraging audio-lyrics contrastive loss and two symmetric cross-modal
attention, to align and fuse features from audio and lyrics. Furthermore, based
on the nature of the multi-label classification, a genre correlations
extraction module is presented to capture and model potential genre
correlations. Extensive experiments demonstrate that our proposed method
significantly surpasses other multi-label music genre classification methods
and achieves state-of-the-art result on Music4All dataset.
| [
{
"version": "v1",
"created": "Tue, 14 Mar 2023 07:16:54 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 01:58:30 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Ru",
"Ganghui",
""
],
[
"Zhang",
"Xulong",
""
],
[
"Wang",
"Jianzong",
""
],
[
"Cheng",
"Ning",
""
],
[
"Xiao",
"Jing",
""
]
] | TITLE: Improving Music Genre Classification from Multi-Modal Properties of
Music and Genre Correlations Perspective
ABSTRACT: Music genre classification has been widely studied in past few years for its
various applications in music information retrieval. Previous works tend to
perform unsatisfactorily, since those methods only use audio content or jointly
use audio content and lyrics content inefficiently. In addition, as genres
normally co-occur in a music track, it is desirable to capture and model the
genre correlations to improve the performance of multi-label music genre
classification. To solve these issues, we present a novel multi-modal method
leveraging audio-lyrics contrastive loss and two symmetric cross-modal
attention, to align and fuse features from audio and lyrics. Furthermore, based
on the nature of the multi-label classification, a genre correlations
extraction module is presented to capture and model potential genre
correlations. Extensive experiments demonstrate that our proposed method
significantly surpasses other multi-label music genre classification methods
and achieves state-of-the-art result on Music4All dataset.
|
2303.10158 | Daochen Zha | Daochen Zha, Zaid Pervaiz Bhat, Kwei-Herng Lai, Fan Yang, Zhimeng
Jiang, Shaochen Zhong, Xia Hu | Data-centric Artificial Intelligence: A Survey | 38 pages, 6 figues, 5 tables. A companion list of data-centric AI
resources is available at https://github.com/daochenzha/data-centric-AI | null | null | null | cs.LG cs.AI cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Artificial Intelligence (AI) is making a profound impact in almost every
domain. A vital enabler of its great success is the availability of abundant
and high-quality data for building machine learning models. Recently, the role
of data in AI has been significantly magnified, giving rise to the emerging
concept of data-centric AI. The attention of researchers and practitioners has
gradually shifted from advancing model design to enhancing the quality and
quantity of the data. In this survey, we discuss the necessity of data-centric
AI, followed by a holistic view of three general data-centric goals (training
data development, inference data development, and data maintenance) and the
representative methods. We also organize the existing literature from
automation and collaboration perspectives, discuss the challenges, and tabulate
the benchmarks for various tasks. We believe this is the first comprehensive
survey that provides a global view of a spectrum of tasks across various stages
of the data lifecycle. We hope it can help the readers efficiently grasp a
broad picture of this field, and equip them with the techniques and further
research ideas to systematically engineer data for building AI systems. A
companion list of data-centric AI resources will be regularly updated on
https://github.com/daochenzha/data-centric-AI
| [
{
"version": "v1",
"created": "Fri, 17 Mar 2023 17:44:56 GMT"
},
{
"version": "v2",
"created": "Sun, 2 Apr 2023 04:43:32 GMT"
},
{
"version": "v3",
"created": "Sun, 11 Jun 2023 07:25:40 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zha",
"Daochen",
""
],
[
"Bhat",
"Zaid Pervaiz",
""
],
[
"Lai",
"Kwei-Herng",
""
],
[
"Yang",
"Fan",
""
],
[
"Jiang",
"Zhimeng",
""
],
[
"Zhong",
"Shaochen",
""
],
[
"Hu",
"Xia",
""
]
] | TITLE: Data-centric Artificial Intelligence: A Survey
ABSTRACT: Artificial Intelligence (AI) is making a profound impact in almost every
domain. A vital enabler of its great success is the availability of abundant
and high-quality data for building machine learning models. Recently, the role
of data in AI has been significantly magnified, giving rise to the emerging
concept of data-centric AI. The attention of researchers and practitioners has
gradually shifted from advancing model design to enhancing the quality and
quantity of the data. In this survey, we discuss the necessity of data-centric
AI, followed by a holistic view of three general data-centric goals (training
data development, inference data development, and data maintenance) and the
representative methods. We also organize the existing literature from
automation and collaboration perspectives, discuss the challenges, and tabulate
the benchmarks for various tasks. We believe this is the first comprehensive
survey that provides a global view of a spectrum of tasks across various stages
of the data lifecycle. We hope it can help the readers efficiently grasp a
broad picture of this field, and equip them with the techniques and further
research ideas to systematically engineer data for building AI systems. A
companion list of data-centric AI resources will be regularly updated on
https://github.com/daochenzha/data-centric-AI
|
2304.06858 | Mohammad Reza Zarei | Mohammad Reza Zarei, Michael Christensen, Sarah Everts and Majid
Komeili | Vax-Culture: A Dataset for Studying Vaccine Discourse on Twitter | null | null | null | null | cs.SI cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Vaccine hesitancy continues to be a main challenge for public health
officials during the COVID-19 pandemic. As this hesitancy undermines vaccine
campaigns, many researchers have sought to identify its root causes, finding
that the increasing volume of anti-vaccine misinformation on social media
platforms is a key element of this problem. We explored Twitter as a source of
misleading content with the goal of extracting overlapping cultural and
political beliefs that motivate the spread of vaccine misinformation. To do
this, we have collected a data set of vaccine-related Tweets and annotated them
with the help of a team of annotators with a background in communications and
journalism. Ultimately we hope this can lead to effective and targeted public
health communication strategies for reaching individuals with anti-vaccine
beliefs. Moreover, this information helps with developing Machine Learning
models to automatically detect vaccine misinformation posts and combat their
negative impacts. In this paper, we present Vax-Culture, a novel Twitter
COVID-19 dataset consisting of 6373 vaccine-related tweets accompanied by an
extensive set of human-provided annotations including vaccine-hesitancy stance,
indication of any misinformation in tweets, the entities criticized and
supported in each tweet and the communicated message of each tweet. Moreover,
we define five baseline tasks including four classification and one sequence
generation tasks, and report the results of a set of recent transformer-based
models for them. The dataset and code are publicly available at
https://github.com/mrzarei5/Vax-Culture.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 23:04:30 GMT"
},
{
"version": "v2",
"created": "Mon, 17 Apr 2023 16:51:08 GMT"
},
{
"version": "v3",
"created": "Sun, 11 Jun 2023 22:11:10 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zarei",
"Mohammad Reza",
""
],
[
"Christensen",
"Michael",
""
],
[
"Everts",
"Sarah",
""
],
[
"Komeili",
"Majid",
""
]
] | TITLE: Vax-Culture: A Dataset for Studying Vaccine Discourse on Twitter
ABSTRACT: Vaccine hesitancy continues to be a main challenge for public health
officials during the COVID-19 pandemic. As this hesitancy undermines vaccine
campaigns, many researchers have sought to identify its root causes, finding
that the increasing volume of anti-vaccine misinformation on social media
platforms is a key element of this problem. We explored Twitter as a source of
misleading content with the goal of extracting overlapping cultural and
political beliefs that motivate the spread of vaccine misinformation. To do
this, we have collected a data set of vaccine-related Tweets and annotated them
with the help of a team of annotators with a background in communications and
journalism. Ultimately we hope this can lead to effective and targeted public
health communication strategies for reaching individuals with anti-vaccine
beliefs. Moreover, this information helps with developing Machine Learning
models to automatically detect vaccine misinformation posts and combat their
negative impacts. In this paper, we present Vax-Culture, a novel Twitter
COVID-19 dataset consisting of 6373 vaccine-related tweets accompanied by an
extensive set of human-provided annotations including vaccine-hesitancy stance,
indication of any misinformation in tweets, the entities criticized and
supported in each tweet and the communicated message of each tweet. Moreover,
we define five baseline tasks including four classification and one sequence
generation tasks, and report the results of a set of recent transformer-based
models for them. The dataset and code are publicly available at
https://github.com/mrzarei5/Vax-Culture.
|
2304.07297 | Hengyuan Hu | Hengyuan Hu, Dorsa Sadigh | Language Instructed Reinforcement Learning for Human-AI Coordination | null | null | null | null | cs.AI cs.CL cs.LG cs.MA | http://creativecommons.org/licenses/by/4.0/ | One of the fundamental quests of AI is to produce agents that coordinate well
with humans. This problem is challenging, especially in domains that lack high
quality human behavioral data, because multi-agent reinforcement learning (RL)
often converges to different equilibria from the ones that humans prefer. We
propose a novel framework, instructRL, that enables humans to specify what kind
of strategies they expect from their AI partners through natural language
instructions. We use pretrained large language models to generate a prior
policy conditioned on the human instruction and use the prior to regularize the
RL objective. This leads to the RL agent converging to equilibria that are
aligned with human preferences. We show that instructRL converges to human-like
policies that satisfy the given instructions in a proof-of-concept environment
as well as the challenging Hanabi benchmark. Finally, we show that knowing the
language instruction significantly boosts human-AI coordination performance in
human evaluations in Hanabi.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 04:47:31 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 20:00:45 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Hu",
"Hengyuan",
""
],
[
"Sadigh",
"Dorsa",
""
]
] | TITLE: Language Instructed Reinforcement Learning for Human-AI Coordination
ABSTRACT: One of the fundamental quests of AI is to produce agents that coordinate well
with humans. This problem is challenging, especially in domains that lack high
quality human behavioral data, because multi-agent reinforcement learning (RL)
often converges to different equilibria from the ones that humans prefer. We
propose a novel framework, instructRL, that enables humans to specify what kind
of strategies they expect from their AI partners through natural language
instructions. We use pretrained large language models to generate a prior
policy conditioned on the human instruction and use the prior to regularize the
RL objective. This leads to the RL agent converging to equilibria that are
aligned with human preferences. We show that instructRL converges to human-like
policies that satisfy the given instructions in a proof-of-concept environment
as well as the challenging Hanabi benchmark. Finally, we show that knowing the
language instruction significantly boosts human-AI coordination performance in
human evaluations in Hanabi.
|
2304.09439 | Dongwon Son | Dongwon Son and Beomjoon Kim | Local object crop collision network for efficient simulation of
non-convex objects in GPU-based simulators | RSS 2023 https://sites.google.com/view/locc-rss2023/home | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | Our goal is to develop an efficient contact detection algorithm for
large-scale GPU-based simulation of non-convex objects. Current GPU-based
simulators such as IsaacGym and Brax must trade-off speed with fidelity,
generality, or both when simulating non-convex objects. Their main issue lies
in contact detection (CD): existing CD algorithms, such as
Gilbert-Johnson-Keerthi (GJK), must trade off their computational speed with
accuracy which becomes expensive as the number of collisions among non-convex
objects increases. We propose a data-driven approach for CD, whose accuracy
depends only on the quality and quantity of offline dataset rather than online
computation time. Unlike GJK, our method inherently has a uniform computational
flow, which facilitates efficient GPU usage based on advanced compilers such as
XLA (Accelerated Linear Algebra). Further, we offer a data-efficient solution
by learning the patterns of colliding local crop object shapes, rather than
global object shapes which are harder to learn. We demonstrate our approach
improves the efficiency of existing CD methods by a factor of 5-10 for
non-convex objects with comparable accuracy. Using the previous work on contact
resolution for a neural-network-based contact detector, we integrate our CD
algorithm into the open-source GPU-based simulator, Brax, and show that we can
improve the efficiency over IsaacGym and generality over standard Brax. We
highly recommend the videos of our simulator included in the supplementary
materials.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 06:09:12 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 14:13:17 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Son",
"Dongwon",
""
],
[
"Kim",
"Beomjoon",
""
]
] | TITLE: Local object crop collision network for efficient simulation of
non-convex objects in GPU-based simulators
ABSTRACT: Our goal is to develop an efficient contact detection algorithm for
large-scale GPU-based simulation of non-convex objects. Current GPU-based
simulators such as IsaacGym and Brax must trade-off speed with fidelity,
generality, or both when simulating non-convex objects. Their main issue lies
in contact detection (CD): existing CD algorithms, such as
Gilbert-Johnson-Keerthi (GJK), must trade off their computational speed with
accuracy which becomes expensive as the number of collisions among non-convex
objects increases. We propose a data-driven approach for CD, whose accuracy
depends only on the quality and quantity of offline dataset rather than online
computation time. Unlike GJK, our method inherently has a uniform computational
flow, which facilitates efficient GPU usage based on advanced compilers such as
XLA (Accelerated Linear Algebra). Further, we offer a data-efficient solution
by learning the patterns of colliding local crop object shapes, rather than
global object shapes which are harder to learn. We demonstrate our approach
improves the efficiency of existing CD methods by a factor of 5-10 for
non-convex objects with comparable accuracy. Using the previous work on contact
resolution for a neural-network-based contact detector, we integrate our CD
algorithm into the open-source GPU-based simulator, Brax, and show that we can
improve the efficiency over IsaacGym and generality over standard Brax. We
highly recommend the videos of our simulator included in the supplementary
materials.
|
2304.10743 | Yuhao Kang | Yuhao Kang and Qianheng Zhang and Robert Roth | The Ethics of AI-Generated Maps: A Study of DALLE 2 and Implications for
Cartography | 9 pages, 3 figures, GIScience 2023 conference | null | null | null | cs.CY cs.HC | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The rapid advancement of artificial intelligence (AI) such as the emergence
of large language models including ChatGPT and DALLE 2 has brought both
opportunities for improving productivity and raised ethical concerns. This
paper investigates the ethics of using artificial intelligence (AI) in
cartography, with a particular focus on the generation of maps using DALLE 2.
To accomplish this, we first create an open-sourced dataset that includes
synthetic (AI-generated) and real-world (human-designed) maps at multiple
scales with a variety settings. We subsequently examine four potential ethical
concerns that may arise from the characteristics of DALLE 2 generated maps,
namely inaccuracies, misleading information, unanticipated features, and
reproducibility. We then develop a deep learning-based ethical examination
system that identifies those AI-generated maps. Our research emphasizes the
importance of ethical considerations in the development and use of AI
techniques in cartography, contributing to the growing body of work on
trustworthy maps. We aim to raise public awareness of the potential risks
associated with AI-generated maps and support the development of ethical
guidelines for their future use.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 04:46:59 GMT"
},
{
"version": "v2",
"created": "Fri, 5 May 2023 19:54:46 GMT"
},
{
"version": "v3",
"created": "Sun, 11 Jun 2023 04:17:35 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Kang",
"Yuhao",
""
],
[
"Zhang",
"Qianheng",
""
],
[
"Roth",
"Robert",
""
]
] | TITLE: The Ethics of AI-Generated Maps: A Study of DALLE 2 and Implications for
Cartography
ABSTRACT: The rapid advancement of artificial intelligence (AI) such as the emergence
of large language models including ChatGPT and DALLE 2 has brought both
opportunities for improving productivity and raised ethical concerns. This
paper investigates the ethics of using artificial intelligence (AI) in
cartography, with a particular focus on the generation of maps using DALLE 2.
To accomplish this, we first create an open-sourced dataset that includes
synthetic (AI-generated) and real-world (human-designed) maps at multiple
scales with a variety settings. We subsequently examine four potential ethical
concerns that may arise from the characteristics of DALLE 2 generated maps,
namely inaccuracies, misleading information, unanticipated features, and
reproducibility. We then develop a deep learning-based ethical examination
system that identifies those AI-generated maps. Our research emphasizes the
importance of ethical considerations in the development and use of AI
techniques in cartography, contributing to the growing body of work on
trustworthy maps. We aim to raise public awareness of the potential risks
associated with AI-generated maps and support the development of ethical
guidelines for their future use.
|
2304.13620 | Raian Rahman | Raian Rahman, Rizvi Hasan, Abdullah Al Farhad, Md Tahmid Rahman
Laskar, Md. Hamjajul Ashmafee, Abu Raihan Mostofa Kamal | ChartSumm: A Comprehensive Benchmark for Automatic Chart Summarization
of Long and Short Summaries | Accepted as a long paper at the Canadian AI 2023 | null | 10.21428/594757db.0b1f96f6 | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Automatic chart to text summarization is an effective tool for the visually
impaired people along with providing precise insights of tabular data in
natural language to the user. A large and well-structured dataset is always a
key part for data driven models. In this paper, we propose ChartSumm: a
large-scale benchmark dataset consisting of a total of 84,363 charts along with
their metadata and descriptions covering a wide range of topics and chart types
to generate short and long summaries. Extensive experiments with strong
baseline models show that even though these models generate fluent and
informative summaries by achieving decent scores in various automatic
evaluation metrics, they often face issues like suffering from hallucination,
missing out important data points, in addition to incorrect explanation of
complex trends in the charts. We also investigated the potential of expanding
ChartSumm to other languages using automated translation tools. These make our
dataset a challenging benchmark for future research.
| [
{
"version": "v1",
"created": "Wed, 26 Apr 2023 15:25:24 GMT"
},
{
"version": "v2",
"created": "Sat, 29 Apr 2023 17:22:08 GMT"
},
{
"version": "v3",
"created": "Sun, 11 Jun 2023 04:07:27 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Rahman",
"Raian",
""
],
[
"Hasan",
"Rizvi",
""
],
[
"Farhad",
"Abdullah Al",
""
],
[
"Laskar",
"Md Tahmid Rahman",
""
],
[
"Ashmafee",
"Md. Hamjajul",
""
],
[
"Kamal",
"Abu Raihan Mostofa",
""
]
] | TITLE: ChartSumm: A Comprehensive Benchmark for Automatic Chart Summarization
of Long and Short Summaries
ABSTRACT: Automatic chart to text summarization is an effective tool for the visually
impaired people along with providing precise insights of tabular data in
natural language to the user. A large and well-structured dataset is always a
key part for data driven models. In this paper, we propose ChartSumm: a
large-scale benchmark dataset consisting of a total of 84,363 charts along with
their metadata and descriptions covering a wide range of topics and chart types
to generate short and long summaries. Extensive experiments with strong
baseline models show that even though these models generate fluent and
informative summaries by achieving decent scores in various automatic
evaluation metrics, they often face issues like suffering from hallucination,
missing out important data points, in addition to incorrect explanation of
complex trends in the charts. We also investigated the potential of expanding
ChartSumm to other languages using automated translation tools. These make our
dataset a challenging benchmark for future research.
|
2304.14836 | Moran Baruch | Moran Baruch, Nir Drucker, Gilad Ezov, Yoav Goldberg, Eyal Kushnir,
Jenny Lerner, Omri Soceanu and Itamar Zimerman | Training Large Scale Polynomial CNNs for E2E Inference over Homomorphic
Encryption | null | null | null | null | cs.LG cs.AI cs.CR | http://creativecommons.org/licenses/by/4.0/ | Training large-scale CNNs that during inference can be run under Homomorphic
Encryption (HE) is challenging due to the need to use only polynomial
operations. This limits HE-based solutions adoption. We address this challenge
and pioneer in providing a novel training method for large polynomial CNNs such
as ResNet-152 and ConvNeXt models, and achieve promising accuracy on encrypted
samples on large-scale dataset such as ImageNet. Additionally, we provide
optimization insights regarding activation functions and skip-connection
latency impacts, enhancing HE-based evaluation efficiency. Finally, to
demonstrate the robustness of our method, we provide a polynomial adaptation of
the CLIP model for secure zero-shot prediction, unlocking unprecedented
capabilities at the intersection of HE and transfer learning.
| [
{
"version": "v1",
"created": "Wed, 26 Apr 2023 20:41:37 GMT"
},
{
"version": "v2",
"created": "Sun, 11 Jun 2023 10:07:52 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Baruch",
"Moran",
""
],
[
"Drucker",
"Nir",
""
],
[
"Ezov",
"Gilad",
""
],
[
"Goldberg",
"Yoav",
""
],
[
"Kushnir",
"Eyal",
""
],
[
"Lerner",
"Jenny",
""
],
[
"Soceanu",
"Omri",
""
],
[
"Zimerman",
"Itamar",
""
]
] | TITLE: Training Large Scale Polynomial CNNs for E2E Inference over Homomorphic
Encryption
ABSTRACT: Training large-scale CNNs that during inference can be run under Homomorphic
Encryption (HE) is challenging due to the need to use only polynomial
operations. This limits HE-based solutions adoption. We address this challenge
and pioneer in providing a novel training method for large polynomial CNNs such
as ResNet-152 and ConvNeXt models, and achieve promising accuracy on encrypted
samples on large-scale dataset such as ImageNet. Additionally, we provide
optimization insights regarding activation functions and skip-connection
latency impacts, enhancing HE-based evaluation efficiency. Finally, to
demonstrate the robustness of our method, we provide a polynomial adaptation of
the CLIP model for secure zero-shot prediction, unlocking unprecedented
capabilities at the intersection of HE and transfer learning.
|
2304.14837 | Fei Xue | Fei Xue and Ignas Budvytis and Roberto Cipolla | IMP: Iterative Matching and Pose Estimation with Adaptive Pooling | CVPR 2023. code available at https://github.com/feixue94/imp-release | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Previous methods solve feature matching and pose estimation using a two-stage
process by first finding matches and then estimating the pose. As they ignore
the geometric relationships between the two tasks, they focus on either
improving the quality of matches or filtering potential outliers, leading to
limited efficiency or accuracy. In contrast, we propose an iterative matching
and pose estimation framework (IMP) leveraging the geometric connections
between the two tasks: a few good matches are enough for a roughly accurate
pose estimation; a roughly accurate pose can be used to guide the matching by
providing geometric constraints. To this end, we implement a geometry-aware
recurrent attention-based module which jointly outputs sparse matches and
camera poses. Specifically, for each iteration, we first implicitly embed
geometric information into the module via a pose-consistency loss, allowing it
to predict geometry-aware matches progressively. Second, we introduce an
\textbf{e}fficient IMP, called EIMP, to dynamically discard keypoints without
potential matches, avoiding redundant updating and significantly reducing the
quadratic time complexity of attention computation in transformers. Experiments
on YFCC100m, Scannet, and Aachen Day-Night datasets demonstrate that the
proposed method outperforms previous approaches in terms of accuracy and
efficiency.
| [
{
"version": "v1",
"created": "Fri, 28 Apr 2023 13:25:50 GMT"
},
{
"version": "v2",
"created": "Sun, 11 Jun 2023 16:31:39 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Xue",
"Fei",
""
],
[
"Budvytis",
"Ignas",
""
],
[
"Cipolla",
"Roberto",
""
]
] | TITLE: IMP: Iterative Matching and Pose Estimation with Adaptive Pooling
ABSTRACT: Previous methods solve feature matching and pose estimation using a two-stage
process by first finding matches and then estimating the pose. As they ignore
the geometric relationships between the two tasks, they focus on either
improving the quality of matches or filtering potential outliers, leading to
limited efficiency or accuracy. In contrast, we propose an iterative matching
and pose estimation framework (IMP) leveraging the geometric connections
between the two tasks: a few good matches are enough for a roughly accurate
pose estimation; a roughly accurate pose can be used to guide the matching by
providing geometric constraints. To this end, we implement a geometry-aware
recurrent attention-based module which jointly outputs sparse matches and
camera poses. Specifically, for each iteration, we first implicitly embed
geometric information into the module via a pose-consistency loss, allowing it
to predict geometry-aware matches progressively. Second, we introduce an
\textbf{e}fficient IMP, called EIMP, to dynamically discard keypoints without
potential matches, avoiding redundant updating and significantly reducing the
quadratic time complexity of attention computation in transformers. Experiments
on YFCC100m, Scannet, and Aachen Day-Night datasets demonstrate that the
proposed method outperforms previous approaches in terms of accuracy and
efficiency.
|
2304.14845 | Fei Xue | Fei Xue and Ignas Budvytis and Roberto Cipolla | SFD2: Semantic-guided Feature Detection and Description | CVPR 2023. code is available at https://github.com/feixue94/sfd2 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual localization is a fundamental task for various applications including
autonomous driving and robotics. Prior methods focus on extracting large
amounts of often redundant locally reliable features, resulting in limited
efficiency and accuracy, especially in large-scale environments under
challenging conditions. Instead, we propose to extract globally reliable
features by implicitly embedding high-level semantics into both the detection
and description processes. Specifically, our semantic-aware detector is able to
detect keypoints from reliable regions (e.g. building, traffic lane) and
suppress unreliable areas (e.g. sky, car) implicitly instead of relying on
explicit semantic labels. This boosts the accuracy of keypoint matching by
reducing the number of features sensitive to appearance changes and avoiding
the need of additional segmentation networks at test time. Moreover, our
descriptors are augmented with semantics and have stronger discriminative
ability, providing more inliers at test time. Particularly, experiments on
long-term large-scale visual localization Aachen Day-Night and RobotCar-Seasons
datasets demonstrate that our model outperforms previous local features and
gives competitive accuracy to advanced matchers but is about 2 and 3 times
faster when using 2k and 4k keypoints, respectively.
| [
{
"version": "v1",
"created": "Fri, 28 Apr 2023 13:34:26 GMT"
},
{
"version": "v2",
"created": "Sun, 11 Jun 2023 16:35:08 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Xue",
"Fei",
""
],
[
"Budvytis",
"Ignas",
""
],
[
"Cipolla",
"Roberto",
""
]
] | TITLE: SFD2: Semantic-guided Feature Detection and Description
ABSTRACT: Visual localization is a fundamental task for various applications including
autonomous driving and robotics. Prior methods focus on extracting large
amounts of often redundant locally reliable features, resulting in limited
efficiency and accuracy, especially in large-scale environments under
challenging conditions. Instead, we propose to extract globally reliable
features by implicitly embedding high-level semantics into both the detection
and description processes. Specifically, our semantic-aware detector is able to
detect keypoints from reliable regions (e.g. building, traffic lane) and
suppress unreliable areas (e.g. sky, car) implicitly instead of relying on
explicit semantic labels. This boosts the accuracy of keypoint matching by
reducing the number of features sensitive to appearance changes and avoiding
the need of additional segmentation networks at test time. Moreover, our
descriptors are augmented with semantics and have stronger discriminative
ability, providing more inliers at test time. Particularly, experiments on
long-term large-scale visual localization Aachen Day-Night and RobotCar-Seasons
datasets demonstrate that our model outperforms previous local features and
gives competitive accuracy to advanced matchers but is about 2 and 3 times
faster when using 2k and 4k keypoints, respectively.
|
2305.00621 | Hiroki Yanagisawa | Hiroki Yanagisawa | Proper Scoring Rules for Survival Analysis | Accepted at ICML 2023 | null | null | null | stat.ME cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Survival analysis is the problem of estimating probability distributions for
future event times, which can be seen as a problem in uncertainty
quantification. Although there are fundamental theories on strictly proper
scoring rules for uncertainty quantification, little is known about those for
survival analysis. In this paper, we investigate extensions of four major
strictly proper scoring rules for survival analysis and we prove that these
extensions are proper under certain conditions, which arise from the
discretization of the estimation of probability distributions. We also compare
the estimation performances of these extended scoring rules by using real
datasets, and the extensions of the logarithmic score and the Brier score
performed the best.
| [
{
"version": "v1",
"created": "Mon, 1 May 2023 01:50:24 GMT"
},
{
"version": "v2",
"created": "Tue, 23 May 2023 07:21:42 GMT"
},
{
"version": "v3",
"created": "Mon, 12 Jun 2023 09:32:29 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Yanagisawa",
"Hiroki",
""
]
] | TITLE: Proper Scoring Rules for Survival Analysis
ABSTRACT: Survival analysis is the problem of estimating probability distributions for
future event times, which can be seen as a problem in uncertainty
quantification. Although there are fundamental theories on strictly proper
scoring rules for uncertainty quantification, little is known about those for
survival analysis. In this paper, we investigate extensions of four major
strictly proper scoring rules for survival analysis and we prove that these
extensions are proper under certain conditions, which arise from the
discretization of the estimation of probability distributions. We also compare
the estimation performances of these extended scoring rules by using real
datasets, and the extensions of the logarithmic score and the Brier score
performed the best.
|
2305.01876 | Siyu Yuan | Siyu Yuan, Deqing Yang, Jinxi Liu, Shuyu Tian, Jiaqing Liang, Yanghua
Xiao, Rui Xie | Causality-aware Concept Extraction based on Knowledge-guided Prompting | Accepted to ACL 2023 | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Concepts benefit natural language understanding but are far from complete in
existing knowledge graphs (KGs). Recently, pre-trained language models (PLMs)
have been widely used in text-based concept extraction (CE). However, PLMs tend
to mine the co-occurrence associations from massive corpus as pre-trained
knowledge rather than the real causal effect between tokens. As a result, the
pre-trained knowledge confounds PLMs to extract biased concepts based on
spurious co-occurrence correlations, inevitably resulting in low precision. In
this paper, through the lens of a Structural Causal Model (SCM), we propose
equipping the PLM-based extractor with a knowledge-guided prompt as an
intervention to alleviate concept bias. The prompt adopts the topic of the
given entity from the existing knowledge in KGs to mitigate the spurious
co-occurrence correlations between entities and biased concepts. Our extensive
experiments on representative multilingual KG datasets justify that our
proposed prompt can effectively alleviate concept bias and improve the
performance of PLM-based CE models.The code has been released on
https://github.com/siyuyuan/KPCE.
| [
{
"version": "v1",
"created": "Wed, 3 May 2023 03:36:20 GMT"
},
{
"version": "v2",
"created": "Thu, 4 May 2023 02:16:38 GMT"
},
{
"version": "v3",
"created": "Sun, 7 May 2023 03:02:12 GMT"
},
{
"version": "v4",
"created": "Wed, 10 May 2023 01:15:45 GMT"
},
{
"version": "v5",
"created": "Sat, 10 Jun 2023 07:34:27 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Yuan",
"Siyu",
""
],
[
"Yang",
"Deqing",
""
],
[
"Liu",
"Jinxi",
""
],
[
"Tian",
"Shuyu",
""
],
[
"Liang",
"Jiaqing",
""
],
[
"Xiao",
"Yanghua",
""
],
[
"Xie",
"Rui",
""
]
] | TITLE: Causality-aware Concept Extraction based on Knowledge-guided Prompting
ABSTRACT: Concepts benefit natural language understanding but are far from complete in
existing knowledge graphs (KGs). Recently, pre-trained language models (PLMs)
have been widely used in text-based concept extraction (CE). However, PLMs tend
to mine the co-occurrence associations from massive corpus as pre-trained
knowledge rather than the real causal effect between tokens. As a result, the
pre-trained knowledge confounds PLMs to extract biased concepts based on
spurious co-occurrence correlations, inevitably resulting in low precision. In
this paper, through the lens of a Structural Causal Model (SCM), we propose
equipping the PLM-based extractor with a knowledge-guided prompt as an
intervention to alleviate concept bias. The prompt adopts the topic of the
given entity from the existing knowledge in KGs to mitigate the spurious
co-occurrence correlations between entities and biased concepts. Our extensive
experiments on representative multilingual KG datasets justify that our
proposed prompt can effectively alleviate concept bias and improve the
performance of PLM-based CE models.The code has been released on
https://github.com/siyuyuan/KPCE.
|
2305.03433 | Kehui Tan | Kehui Tan, Tianqi Pang, Chenyou Fan and Song Yu | Towards Applying Powerful Large AI Models in Classroom Teaching:
Opportunities, Challenges and Prospects | 16 pages, 2 figures | null | null | null | cs.AI cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This perspective paper proposes a series of interactive scenarios that
utilize Artificial Intelligence (AI) to enhance classroom teaching, such as
dialogue auto-completion, knowledge and style transfer, and assessment of
AI-generated content. By leveraging recent developments in Large Language
Models (LLMs), we explore the potential of AI to augment and enrich
teacher-student dialogues and improve the quality of teaching. Our goal is to
produce innovative and meaningful conversations between teachers and students,
create standards for evaluation, and improve the efficacy of AI-for-Education
initiatives. In Section 3, we discuss the challenges of utilizing existing LLMs
to effectively complete the educated tasks and present a unified framework for
addressing diverse education dataset, processing lengthy conversations, and
condensing information to better accomplish more downstream tasks. In Section
4, we summarize the pivoting tasks including Teacher-Student Dialogue
Auto-Completion, Expert Teaching Knowledge and Style Transfer, and Assessment
of AI-Generated Content (AIGC), providing a clear path for future research. In
Section 5, we also explore the use of external and adjustable LLMs to improve
the generated content through human-in-the-loop supervision and reinforcement
learning. Ultimately, this paper seeks to highlight the potential for AI to aid
the field of education and promote its further exploration.
| [
{
"version": "v1",
"created": "Fri, 5 May 2023 11:09:13 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 11:53:37 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Tan",
"Kehui",
""
],
[
"Pang",
"Tianqi",
""
],
[
"Fan",
"Chenyou",
""
],
[
"Yu",
"Song",
""
]
] | TITLE: Towards Applying Powerful Large AI Models in Classroom Teaching:
Opportunities, Challenges and Prospects
ABSTRACT: This perspective paper proposes a series of interactive scenarios that
utilize Artificial Intelligence (AI) to enhance classroom teaching, such as
dialogue auto-completion, knowledge and style transfer, and assessment of
AI-generated content. By leveraging recent developments in Large Language
Models (LLMs), we explore the potential of AI to augment and enrich
teacher-student dialogues and improve the quality of teaching. Our goal is to
produce innovative and meaningful conversations between teachers and students,
create standards for evaluation, and improve the efficacy of AI-for-Education
initiatives. In Section 3, we discuss the challenges of utilizing existing LLMs
to effectively complete the educated tasks and present a unified framework for
addressing diverse education dataset, processing lengthy conversations, and
condensing information to better accomplish more downstream tasks. In Section
4, we summarize the pivoting tasks including Teacher-Student Dialogue
Auto-Completion, Expert Teaching Knowledge and Style Transfer, and Assessment
of AI-Generated Content (AIGC), providing a clear path for future research. In
Section 5, we also explore the use of external and adjustable LLMs to improve
the generated content through human-in-the-loop supervision and reinforcement
learning. Ultimately, this paper seeks to highlight the potential for AI to aid
the field of education and promote its further exploration.
|
2305.09779 | Ali Gorji | Ali Gorji, Andisheh Amrollahi, Andreas Krause | A Scalable Walsh-Hadamard Regularizer to Overcome the Low-degree
Spectral Bias of Neural Networks | Accepted for the 39th Conference on Uncertainty in Artificial
Intelligence (UAI 2023) | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Despite the capacity of neural nets to learn arbitrary functions, models
trained through gradient descent often exhibit a bias towards ``simpler''
functions. Various notions of simplicity have been introduced to characterize
this behavior. Here, we focus on the case of neural networks with discrete
(zero-one), high-dimensional, inputs through the lens of their Fourier
(Walsh-Hadamard) transforms, where the notion of simplicity can be captured
through the degree of the Fourier coefficients. We empirically show that neural
networks have a tendency to learn lower-degree frequencies. We show how this
spectral bias towards low-degree frequencies can in fact hurt the neural
network's generalization on real-world datasets. To remedy this we propose a
new scalable functional regularization scheme that aids the neural network to
learn higher degree frequencies. Our regularizer also helps avoid erroneous
identification of low-degree frequencies, which further improves
generalization. We extensively evaluate our regularizer on synthetic datasets
to gain insights into its behavior. Finally, we show significantly improved
generalization on four different datasets compared to standard neural networks
and other relevant baselines.
| [
{
"version": "v1",
"created": "Tue, 16 May 2023 20:06:01 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 09:10:14 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Gorji",
"Ali",
""
],
[
"Amrollahi",
"Andisheh",
""
],
[
"Krause",
"Andreas",
""
]
] | TITLE: A Scalable Walsh-Hadamard Regularizer to Overcome the Low-degree
Spectral Bias of Neural Networks
ABSTRACT: Despite the capacity of neural nets to learn arbitrary functions, models
trained through gradient descent often exhibit a bias towards ``simpler''
functions. Various notions of simplicity have been introduced to characterize
this behavior. Here, we focus on the case of neural networks with discrete
(zero-one), high-dimensional, inputs through the lens of their Fourier
(Walsh-Hadamard) transforms, where the notion of simplicity can be captured
through the degree of the Fourier coefficients. We empirically show that neural
networks have a tendency to learn lower-degree frequencies. We show how this
spectral bias towards low-degree frequencies can in fact hurt the neural
network's generalization on real-world datasets. To remedy this we propose a
new scalable functional regularization scheme that aids the neural network to
learn higher degree frequencies. Our regularizer also helps avoid erroneous
identification of low-degree frequencies, which further improves
generalization. We extensively evaluate our regularizer on synthetic datasets
to gain insights into its behavior. Finally, we show significantly improved
generalization on four different datasets compared to standard neural networks
and other relevant baselines.
|
2305.12831 | Yidi Jiang | Yidi Jiang, Ruijie Tao, Zexu Pan, Haizhou Li | Target Active Speaker Detection with Audio-visual Cues | Accepted to INTERSPEECH2023 | null | null | null | eess.AS cs.SD | http://creativecommons.org/licenses/by/4.0/ | In active speaker detection (ASD), we would like to detect whether an
on-screen person is speaking based on audio-visual cues. Previous studies have
primarily focused on modeling audio-visual synchronization cue, which depends
on the video quality of the lip region of a speaker. In real-world
applications, it is possible that we can also have the reference speech of the
on-screen speaker. To benefit from both facial cue and reference speech, we
propose the Target Speaker TalkNet (TS-TalkNet), which leverages a pre-enrolled
speaker embedding to complement the audio-visual synchronization cue in
detecting whether the target speaker is speaking. Our framework outperforms the
popular model, TalkNet on two datasets, achieving absolute improvements of 1.6%
in mAP on the AVA-ActiveSpeaker validation set, and 0.8%, 0.4%, and 0.8% in
terms of AP, AUC and EER on the ASW test set, respectively. Code is available
at https://github.com/Jiang-Yidi/TS-TalkNet/.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 08:52:42 GMT"
},
{
"version": "v2",
"created": "Fri, 26 May 2023 08:54:48 GMT"
},
{
"version": "v3",
"created": "Mon, 12 Jun 2023 06:41:17 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Jiang",
"Yidi",
""
],
[
"Tao",
"Ruijie",
""
],
[
"Pan",
"Zexu",
""
],
[
"Li",
"Haizhou",
""
]
] | TITLE: Target Active Speaker Detection with Audio-visual Cues
ABSTRACT: In active speaker detection (ASD), we would like to detect whether an
on-screen person is speaking based on audio-visual cues. Previous studies have
primarily focused on modeling audio-visual synchronization cue, which depends
on the video quality of the lip region of a speaker. In real-world
applications, it is possible that we can also have the reference speech of the
on-screen speaker. To benefit from both facial cue and reference speech, we
propose the Target Speaker TalkNet (TS-TalkNet), which leverages a pre-enrolled
speaker embedding to complement the audio-visual synchronization cue in
detecting whether the target speaker is speaking. Our framework outperforms the
popular model, TalkNet on two datasets, achieving absolute improvements of 1.6%
in mAP on the AVA-ActiveSpeaker validation set, and 0.8%, 0.4%, and 0.8% in
terms of AP, AUC and EER on the ASW test set, respectively. Code is available
at https://github.com/Jiang-Yidi/TS-TalkNet/.
|
2305.13119 | Zhu Liu | Zhu Liu and Ying Liu | Ambiguity Meets Uncertainty: Investigating Uncertainty Estimation for
Word Sense Disambiguation | Findings: ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Word sense disambiguation (WSD), which aims to determine an appropriate sense
for a target word given its context, is crucial for natural language
understanding. Existing supervised methods treat WSD as a classification task
and have achieved remarkable performance. However, they ignore uncertainty
estimation (UE) in the real-world setting, where the data is always noisy and
out of distribution. This paper extensively studies UE on the benchmark
designed for WSD. Specifically, we first compare four uncertainty scores for a
state-of-the-art WSD model and verify that the conventional predictive
probabilities obtained at the end of the model are inadequate to quantify
uncertainty. Then, we examine the capability of capturing data and model
uncertainties by the model with the selected UE score on well-designed test
scenarios and discover that the model reflects data uncertainty satisfactorily
but underestimates model uncertainty. Furthermore, we explore numerous lexical
properties that intrinsically affect data uncertainty and provide a detailed
analysis of four critical aspects: the syntactic category, morphology, sense
granularity, and semantic relations. The code is available at
https://github.com/RyanLiut/WSD-UE.
| [
{
"version": "v1",
"created": "Mon, 22 May 2023 15:18:15 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 03:27:31 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Liu",
"Zhu",
""
],
[
"Liu",
"Ying",
""
]
] | TITLE: Ambiguity Meets Uncertainty: Investigating Uncertainty Estimation for
Word Sense Disambiguation
ABSTRACT: Word sense disambiguation (WSD), which aims to determine an appropriate sense
for a target word given its context, is crucial for natural language
understanding. Existing supervised methods treat WSD as a classification task
and have achieved remarkable performance. However, they ignore uncertainty
estimation (UE) in the real-world setting, where the data is always noisy and
out of distribution. This paper extensively studies UE on the benchmark
designed for WSD. Specifically, we first compare four uncertainty scores for a
state-of-the-art WSD model and verify that the conventional predictive
probabilities obtained at the end of the model are inadequate to quantify
uncertainty. Then, we examine the capability of capturing data and model
uncertainties by the model with the selected UE score on well-designed test
scenarios and discover that the model reflects data uncertainty satisfactorily
but underestimates model uncertainty. Furthermore, we explore numerous lexical
properties that intrinsically affect data uncertainty and provide a detailed
analysis of four critical aspects: the syntactic category, morphology, sense
granularity, and semantic relations. The code is available at
https://github.com/RyanLiut/WSD-UE.
|
2305.15676 | Yuejiao Fei | Yuejiao Fei, Leyang Cui, Sen Yang, Wai Lam, Zhenzhong Lan, Shuming Shi | Enhancing Grammatical Error Correction Systems with Explanations | 9 pages, 7 figures, accepted to the main conference of ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Grammatical error correction systems improve written communication by
detecting and correcting language mistakes. To help language learners better
understand why the GEC system makes a certain correction, the causes of errors
(evidence words) and the corresponding error types are two key factors. To
enhance GEC systems with explanations, we introduce EXPECT, a large dataset
annotated with evidence words and grammatical error types. We propose several
baselines and analysis to understand this task. Furthermore, human evaluation
verifies our explainable GEC system's explanations can assist second-language
learners in determining whether to accept a correction suggestion and in
understanding the associated grammar rule.
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 03:00:49 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 05:41:34 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Fei",
"Yuejiao",
""
],
[
"Cui",
"Leyang",
""
],
[
"Yang",
"Sen",
""
],
[
"Lam",
"Wai",
""
],
[
"Lan",
"Zhenzhong",
""
],
[
"Shi",
"Shuming",
""
]
] | TITLE: Enhancing Grammatical Error Correction Systems with Explanations
ABSTRACT: Grammatical error correction systems improve written communication by
detecting and correcting language mistakes. To help language learners better
understand why the GEC system makes a certain correction, the causes of errors
(evidence words) and the corresponding error types are two key factors. To
enhance GEC systems with explanations, we introduce EXPECT, a large dataset
annotated with evidence words and grammatical error types. We propose several
baselines and analysis to understand this task. Furthermore, human evaluation
verifies our explainable GEC system's explanations can assist second-language
learners in determining whether to accept a correction suggestion and in
understanding the associated grammar rule.
|
2305.16588 | Jie Sun | Jie Sun, Li Su, Zuocheng Shi, Wenting Shen, Zeke Wang, Lei Wang, Jie
Zhang, Yong Li, Wenyuan Yu, Jingren Zhou, Fei Wu | Legion: Automatically Pushing the Envelope of Multi-GPU System for
Billion-Scale GNN Training | null | null | null | null | cs.DC | http://creativecommons.org/licenses/by/4.0/ | Graph neural network(GNN) has been widely applied in real-world applications,
such as product recommendation in e-commerce platforms and risk control in
financial management systems. Several cache-based GNN systems have been built
to accelerate GNN training in a single machine with multiple GPUs. However,
these systems fail to train billion-scale graphs efficiently, which is a common
challenge in the industry. In this work, we propose Legion, a system that
automatically pushes the envelope of multi-GPU systems for accelerating
billion-scale GNN training. First, we design a hierarchical graph partitioning
mechanism that significantly improves the multi-GPU cache performance. Second,
we build a unified multi-GPU cache that helps to minimize the PCIe traffic
incurred by caching both graph topology and features with the highest hotness.
Third, we develop an automatic caching management mechanism that adapts the
multi-GPU cache plan according to the hardware specifications and various
graphs to maximize the overall training throughput. Evaluations on various GNN
models and multiple datasets show that Legion supports training billion-scale
GNNs in a single machine and significantly outperforms the state-of-the-art
cache-based systems on small graphs.
| [
{
"version": "v1",
"created": "Fri, 26 May 2023 02:30:38 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 08:25:09 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Sun",
"Jie",
""
],
[
"Su",
"Li",
""
],
[
"Shi",
"Zuocheng",
""
],
[
"Shen",
"Wenting",
""
],
[
"Wang",
"Zeke",
""
],
[
"Wang",
"Lei",
""
],
[
"Zhang",
"Jie",
""
],
[
"Li",
"Yong",
""
],
[
"Yu",
"Wenyuan",
""
],
[
"Zhou",
"Jingren",
""
],
[
"Wu",
"Fei",
""
]
] | TITLE: Legion: Automatically Pushing the Envelope of Multi-GPU System for
Billion-Scale GNN Training
ABSTRACT: Graph neural network(GNN) has been widely applied in real-world applications,
such as product recommendation in e-commerce platforms and risk control in
financial management systems. Several cache-based GNN systems have been built
to accelerate GNN training in a single machine with multiple GPUs. However,
these systems fail to train billion-scale graphs efficiently, which is a common
challenge in the industry. In this work, we propose Legion, a system that
automatically pushes the envelope of multi-GPU systems for accelerating
billion-scale GNN training. First, we design a hierarchical graph partitioning
mechanism that significantly improves the multi-GPU cache performance. Second,
we build a unified multi-GPU cache that helps to minimize the PCIe traffic
incurred by caching both graph topology and features with the highest hotness.
Third, we develop an automatic caching management mechanism that adapts the
multi-GPU cache plan according to the hardware specifications and various
graphs to maximize the overall training throughput. Evaluations on various GNN
models and multiple datasets show that Legion supports training billion-scale
GNNs in a single machine and significantly outperforms the state-of-the-art
cache-based systems on small graphs.
|
2305.16967 | Bohao Yang | Kun Zhao, Bohao Yang, Chenghua Lin, Wenge Rong, Aline Villavicencio
and Xiaohui Cui | Evaluating Open-Domain Dialogues in Latent Space with Next Sentence
Prediction and Mutual Information | Accepted at ACL2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The long-standing one-to-many issue of the open-domain dialogues poses
significant challenges for automatic evaluation methods, i.e., there may be
multiple suitable responses which differ in semantics for a given
conversational context. To tackle this challenge, we propose a novel
learning-based automatic evaluation metric (CMN), which can robustly evaluate
open-domain dialogues by augmenting Conditional Variational Autoencoders
(CVAEs) with a Next Sentence Prediction (NSP) objective and employing Mutual
Information (MI) to model the semantic similarity of text in the latent space.
Experimental results on two open-domain dialogue datasets demonstrate the
superiority of our method compared with a wide range of baselines, especially
in handling responses which are distant to the golden reference responses in
semantics.
| [
{
"version": "v1",
"created": "Fri, 26 May 2023 14:21:54 GMT"
},
{
"version": "v2",
"created": "Tue, 30 May 2023 15:25:13 GMT"
},
{
"version": "v3",
"created": "Sat, 10 Jun 2023 13:23:41 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zhao",
"Kun",
""
],
[
"Yang",
"Bohao",
""
],
[
"Lin",
"Chenghua",
""
],
[
"Rong",
"Wenge",
""
],
[
"Villavicencio",
"Aline",
""
],
[
"Cui",
"Xiaohui",
""
]
] | TITLE: Evaluating Open-Domain Dialogues in Latent Space with Next Sentence
Prediction and Mutual Information
ABSTRACT: The long-standing one-to-many issue of the open-domain dialogues poses
significant challenges for automatic evaluation methods, i.e., there may be
multiple suitable responses which differ in semantics for a given
conversational context. To tackle this challenge, we propose a novel
learning-based automatic evaluation metric (CMN), which can robustly evaluate
open-domain dialogues by augmenting Conditional Variational Autoencoders
(CVAEs) with a Next Sentence Prediction (NSP) objective and employing Mutual
Information (MI) to model the semantic similarity of text in the latent space.
Experimental results on two open-domain dialogue datasets demonstrate the
superiority of our method compared with a wide range of baselines, especially
in handling responses which are distant to the golden reference responses in
semantics.
|
2305.17537 | Andrey Kurenkov | Andrey Kurenkov, Michael Lingelbach, Tanmay Agarwal, Emily Jin,
Chengshu Li, Ruohan Zhang, Li Fei-Fei, Jiajun Wu, Silvio Savarese, Roberto
Mart\'in-Mart\'in | Modeling Dynamic Environments with Scene Graph Memory | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Embodied AI agents that search for objects in large environments such as
households often need to make efficient decisions by predicting object
locations based on partial information. We pose this as a new type of link
prediction problem: link prediction on partially observable dynamic graphs. Our
graph is a representation of a scene in which rooms and objects are nodes, and
their relationships are encoded in the edges; only parts of the changing graph
are known to the agent at each timestep. This partial observability poses a
challenge to existing link prediction approaches, which we address. We propose
a novel state representation -- Scene Graph Memory (SGM) -- with captures the
agent's accumulated set of observations, as well as a neural net architecture
called a Node Edge Predictor (NEP) that extracts information from the SGM to
search efficiently. We evaluate our method in the Dynamic House Simulator, a
new benchmark that creates diverse dynamic graphs following the semantic
patterns typically seen at homes, and show that NEP can be trained to predict
the locations of objects in a variety of environments with diverse object
movement dynamics, outperforming baselines both in terms of new scene
adaptability and overall accuracy. The codebase and more can be found at
https://www.scenegraphmemory.com.
| [
{
"version": "v1",
"created": "Sat, 27 May 2023 17:39:38 GMT"
},
{
"version": "v2",
"created": "Wed, 31 May 2023 10:33:11 GMT"
},
{
"version": "v3",
"created": "Fri, 2 Jun 2023 21:49:22 GMT"
},
{
"version": "v4",
"created": "Mon, 12 Jun 2023 17:25:06 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Kurenkov",
"Andrey",
""
],
[
"Lingelbach",
"Michael",
""
],
[
"Agarwal",
"Tanmay",
""
],
[
"Jin",
"Emily",
""
],
[
"Li",
"Chengshu",
""
],
[
"Zhang",
"Ruohan",
""
],
[
"Fei-Fei",
"Li",
""
],
[
"Wu",
"Jiajun",
""
],
[
"Savarese",
"Silvio",
""
],
[
"Martín-Martín",
"Roberto",
""
]
] | TITLE: Modeling Dynamic Environments with Scene Graph Memory
ABSTRACT: Embodied AI agents that search for objects in large environments such as
households often need to make efficient decisions by predicting object
locations based on partial information. We pose this as a new type of link
prediction problem: link prediction on partially observable dynamic graphs. Our
graph is a representation of a scene in which rooms and objects are nodes, and
their relationships are encoded in the edges; only parts of the changing graph
are known to the agent at each timestep. This partial observability poses a
challenge to existing link prediction approaches, which we address. We propose
a novel state representation -- Scene Graph Memory (SGM) -- with captures the
agent's accumulated set of observations, as well as a neural net architecture
called a Node Edge Predictor (NEP) that extracts information from the SGM to
search efficiently. We evaluate our method in the Dynamic House Simulator, a
new benchmark that creates diverse dynamic graphs following the semantic
patterns typically seen at homes, and show that NEP can be trained to predict
the locations of objects in a variety of environments with diverse object
movement dynamics, outperforming baselines both in terms of new scene
adaptability and overall accuracy. The codebase and more can be found at
https://www.scenegraphmemory.com.
|
2305.19344 | Jaehyung Kim | Jaehyung Kim, Yekyung Kim, Karin de Langis, Jinwoo Shin, Dongyeop Kang | infoVerse: A Universal Framework for Dataset Characterization with
Multidimensional Meta-information | 26 pages, accepted at ACL 2023 conference as a full paper | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The success of NLP systems often relies on the availability of large,
high-quality datasets. However, not all samples in these datasets are equally
valuable for learning, as some may be redundant or noisy. Several methods for
characterizing datasets based on model-driven meta-information (e.g., model's
confidence) have been developed, but the relationship and complementary effects
of these methods have received less attention. In this paper, we introduce
infoVerse, a universal framework for dataset characterization, which provides a
new feature space that effectively captures multidimensional characteristics of
datasets by incorporating various model-driven meta-information. infoVerse
reveals distinctive regions of the dataset that are not apparent in the
original semantic space, hence guiding users (or models) in identifying which
samples to focus on for exploration, assessment, or annotation. Additionally,
we propose a novel sampling method on infoVerse to select a set of data points
that maximizes informativeness. In three real-world applications (data pruning,
active learning, and data annotation), the samples chosen on infoVerse space
consistently outperform strong baselines in all applications. Our code and demo
are publicly available.
| [
{
"version": "v1",
"created": "Tue, 30 May 2023 18:12:48 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 10:46:10 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Kim",
"Jaehyung",
""
],
[
"Kim",
"Yekyung",
""
],
[
"de Langis",
"Karin",
""
],
[
"Shin",
"Jinwoo",
""
],
[
"Kang",
"Dongyeop",
""
]
] | TITLE: infoVerse: A Universal Framework for Dataset Characterization with
Multidimensional Meta-information
ABSTRACT: The success of NLP systems often relies on the availability of large,
high-quality datasets. However, not all samples in these datasets are equally
valuable for learning, as some may be redundant or noisy. Several methods for
characterizing datasets based on model-driven meta-information (e.g., model's
confidence) have been developed, but the relationship and complementary effects
of these methods have received less attention. In this paper, we introduce
infoVerse, a universal framework for dataset characterization, which provides a
new feature space that effectively captures multidimensional characteristics of
datasets by incorporating various model-driven meta-information. infoVerse
reveals distinctive regions of the dataset that are not apparent in the
original semantic space, hence guiding users (or models) in identifying which
samples to focus on for exploration, assessment, or annotation. Additionally,
we propose a novel sampling method on infoVerse to select a set of data points
that maximizes informativeness. In three real-world applications (data pruning,
active learning, and data annotation), the samples chosen on infoVerse space
consistently outperform strong baselines in all applications. Our code and demo
are publicly available.
|
2305.19512 | Yiwei Lyu | Yiwei Lyu, Tiange Luo, Jiacheng Shi, Todd C. Hollon, Honglak Lee | Fine-grained Text Style Transfer with Diffusion-Based Language Models | Accepted at Repl4NLP workshop at ACL 2023 | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Diffusion probabilistic models have shown great success in generating
high-quality images controllably, and researchers have tried to utilize this
controllability into text generation domain. Previous works on diffusion-based
language models have shown that they can be trained without external knowledge
(such as pre-trained weights) and still achieve stable performance and
controllability. In this paper, we trained a diffusion-based model on StylePTB
dataset, the standard benchmark for fine-grained text style transfers. The
tasks in StylePTB requires much more refined control over the output text
compared to tasks evaluated in previous works, and our model was able to
achieve state-of-the-art performance on StylePTB on both individual and
compositional transfers. Moreover, our model, trained on limited data from
StylePTB without external knowledge, outperforms previous works that utilized
pretrained weights, embeddings, and external grammar parsers, and this may
indicate that diffusion-based language models have great potential under
low-resource settings.
| [
{
"version": "v1",
"created": "Wed, 31 May 2023 02:51:26 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 02:13:16 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Lyu",
"Yiwei",
""
],
[
"Luo",
"Tiange",
""
],
[
"Shi",
"Jiacheng",
""
],
[
"Hollon",
"Todd C.",
""
],
[
"Lee",
"Honglak",
""
]
] | TITLE: Fine-grained Text Style Transfer with Diffusion-Based Language Models
ABSTRACT: Diffusion probabilistic models have shown great success in generating
high-quality images controllably, and researchers have tried to utilize this
controllability into text generation domain. Previous works on diffusion-based
language models have shown that they can be trained without external knowledge
(such as pre-trained weights) and still achieve stable performance and
controllability. In this paper, we trained a diffusion-based model on StylePTB
dataset, the standard benchmark for fine-grained text style transfers. The
tasks in StylePTB requires much more refined control over the output text
compared to tasks evaluated in previous works, and our model was able to
achieve state-of-the-art performance on StylePTB on both individual and
compositional transfers. Moreover, our model, trained on limited data from
StylePTB without external knowledge, outperforms previous works that utilized
pretrained weights, embeddings, and external grammar parsers, and this may
indicate that diffusion-based language models have great potential under
low-resource settings.
|
2305.19903 | Kai-Xuan Chen | Kaixuan Chen and Shunyu Liu and Tongtian Zhu and Tongya Zheng and
Haofei Zhang and Zunlei Feng and Jingwen Ye and Mingli Song | Improving Expressivity of GNNs with Subgraph-specific Factor Embedded
Normalization | 13 pages, 7 figures | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph Neural Networks (GNNs) have emerged as a powerful category of learning
architecture for handling graph-structured data. However, existing GNNs
typically ignore crucial structural characteristics in node-induced subgraphs,
which thus limits their expressiveness for various downstream tasks. In this
paper, we strive to strengthen the representative capabilities of GNNs by
devising a dedicated plug-and-play normalization scheme, termed as
SUbgraph-sPEcific FactoR Embedded Normalization (SuperNorm), that explicitly
considers the intra-connection information within each node-induced subgraph.
To this end, we embed the subgraph-specific factor at the beginning and the end
of the standard BatchNorm, as well as incorporate graph instance-specific
statistics for improved distinguishable capabilities. In the meantime, we
provide theoretical analysis to support that, with the elaborated SuperNorm, an
arbitrary GNN is at least as powerful as the 1-WL test in distinguishing
non-isomorphism graphs. Furthermore, the proposed SuperNorm scheme is also
demonstrated to alleviate the over-smoothing phenomenon. Experimental results
related to predictions of graph, node, and link properties on the eight popular
datasets demonstrate the effectiveness of the proposed method. The code is
available at https://github.com/chenchkx/SuperNorm.
| [
{
"version": "v1",
"created": "Wed, 31 May 2023 14:37:31 GMT"
},
{
"version": "v2",
"created": "Wed, 7 Jun 2023 13:07:52 GMT"
},
{
"version": "v3",
"created": "Sun, 11 Jun 2023 16:19:56 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Chen",
"Kaixuan",
""
],
[
"Liu",
"Shunyu",
""
],
[
"Zhu",
"Tongtian",
""
],
[
"Zheng",
"Tongya",
""
],
[
"Zhang",
"Haofei",
""
],
[
"Feng",
"Zunlei",
""
],
[
"Ye",
"Jingwen",
""
],
[
"Song",
"Mingli",
""
]
] | TITLE: Improving Expressivity of GNNs with Subgraph-specific Factor Embedded
Normalization
ABSTRACT: Graph Neural Networks (GNNs) have emerged as a powerful category of learning
architecture for handling graph-structured data. However, existing GNNs
typically ignore crucial structural characteristics in node-induced subgraphs,
which thus limits their expressiveness for various downstream tasks. In this
paper, we strive to strengthen the representative capabilities of GNNs by
devising a dedicated plug-and-play normalization scheme, termed as
SUbgraph-sPEcific FactoR Embedded Normalization (SuperNorm), that explicitly
considers the intra-connection information within each node-induced subgraph.
To this end, we embed the subgraph-specific factor at the beginning and the end
of the standard BatchNorm, as well as incorporate graph instance-specific
statistics for improved distinguishable capabilities. In the meantime, we
provide theoretical analysis to support that, with the elaborated SuperNorm, an
arbitrary GNN is at least as powerful as the 1-WL test in distinguishing
non-isomorphism graphs. Furthermore, the proposed SuperNorm scheme is also
demonstrated to alleviate the over-smoothing phenomenon. Experimental results
related to predictions of graph, node, and link properties on the eight popular
datasets demonstrate the effectiveness of the proposed method. The code is
available at https://github.com/chenchkx/SuperNorm.
|
2306.00369 | Congda Ma | Congda Ma, Tianyu Zhao, Makoto Shing, Kei Sawada, Manabu Okumura | Focused Prefix Tuning for Controllable Text Generation | Accepted to the ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In a controllable text generation dataset, there exist unannotated attributes
that could provide irrelevant learning signals to models that use it for
training and thus degrade their performance. We propose focused prefix
tuning(FPT) to mitigate the problem and to enable the control to focus on the
desired attribute. Experimental results show that FPT can achieve better
control accuracy and text fluency than baseline models in single-attribute
control tasks. In multi-attribute control tasks, FPT achieves comparable
control accuracy with the state-of-the-art approach while keeping the
flexibility to control new attributes without retraining existing models.
| [
{
"version": "v1",
"created": "Thu, 1 Jun 2023 06:00:43 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 12:36:48 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Ma",
"Congda",
""
],
[
"Zhao",
"Tianyu",
""
],
[
"Shing",
"Makoto",
""
],
[
"Sawada",
"Kei",
""
],
[
"Okumura",
"Manabu",
""
]
] | TITLE: Focused Prefix Tuning for Controllable Text Generation
ABSTRACT: In a controllable text generation dataset, there exist unannotated attributes
that could provide irrelevant learning signals to models that use it for
training and thus degrade their performance. We propose focused prefix
tuning(FPT) to mitigate the problem and to enable the control to focus on the
desired attribute. Experimental results show that FPT can achieve better
control accuracy and text fluency than baseline models in single-attribute
control tasks. In multi-attribute control tasks, FPT achieves comparable
control accuracy with the state-of-the-art approach while keeping the
flexibility to control new attributes without retraining existing models.
|
2306.02395 | Ganlin Zhang | Ganlin Zhang, Deheng Zhang, Feichi Lu, Anqi Li | NICE-SLAM with Adaptive Feature Grids | This is a course project, not suitable for a preprint platform | null | null | null | cs.CV cs.GR | http://creativecommons.org/licenses/by/4.0/ | NICE-SLAM is a dense visual SLAM system that combines the advantages of
neural implicit representations and hierarchical grid-based scene
representation. However, the hierarchical grid features are densely stored,
leading to memory explosion problems when adapting the framework to large
scenes. In our project, we present sparse NICE-SLAM, a sparse SLAM system
incorporating the idea of Voxel Hashing into NICE-SLAM framework. Instead of
initializing feature grids in the whole space, voxel features near the surface
are adaptively added and optimized. Experiments demonstrated that compared to
NICE-SLAM algorithm, our approach takes much less memory and achieves
comparable reconstruction quality on the same datasets. Our implementation is
available at
https://github.com/zhangganlin/NICE-SLAM-with-Adaptive-Feature-Grids.
| [
{
"version": "v1",
"created": "Sun, 4 Jun 2023 16:11:45 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 06:13:54 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zhang",
"Ganlin",
""
],
[
"Zhang",
"Deheng",
""
],
[
"Lu",
"Feichi",
""
],
[
"Li",
"Anqi",
""
]
] | TITLE: NICE-SLAM with Adaptive Feature Grids
ABSTRACT: NICE-SLAM is a dense visual SLAM system that combines the advantages of
neural implicit representations and hierarchical grid-based scene
representation. However, the hierarchical grid features are densely stored,
leading to memory explosion problems when adapting the framework to large
scenes. In our project, we present sparse NICE-SLAM, a sparse SLAM system
incorporating the idea of Voxel Hashing into NICE-SLAM framework. Instead of
initializing feature grids in the whole space, voxel features near the surface
are adaptively added and optimized. Experiments demonstrated that compared to
NICE-SLAM algorithm, our approach takes much less memory and achieves
comparable reconstruction quality on the same datasets. Our implementation is
available at
https://github.com/zhangganlin/NICE-SLAM-with-Adaptive-Feature-Grids.
|
2306.02517 | Takato Yasuno | Takato Yasuno, Masahiro Okano and Junichiro Fujii | Disaster Anomaly Detector via Deeper FCDDs for Explainable Initial
Responses | 10 pages, 10 figures, 8 tables | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Extreme natural disasters can have devastating effects on both urban and
rural areas. In any disaster event, an initial response is the key to rescue
within 72 hours and prompt recovery. During the initial stage of disaster
response, it is important to quickly assess the damage over a wide area and
identify priority areas. Among machine learning algorithms, deep anomaly
detection is effective in detecting devastation features that are different
from everyday features. In addition, explainable computer vision applications
should justify the initial responses. In this paper, we propose an anomaly
detection application utilizing deeper fully convolutional data descriptions
(FCDDs), that enables the localization of devastation features and
visualization of damage-marked heatmaps. More specifically, we show numerous
training and test results for a dataset AIDER with the four disaster
categories: collapsed buildings, traffic incidents, fires, and flooded areas.
We also implement ablation studies of anomalous class imbalance and the data
scale competing against the normal class. Our experiments provide results of
high accuracies over 95% for F1. Furthermore, we found that the deeper FCDD
with a VGG16 backbone consistently outperformed other baselines CNN27,
ResNet101, and Inceptionv3. This study presents a new solution that offers a
disaster anomaly detection application for initial responses with higher
accuracy and devastation explainability, providing a novel contribution to the
prompt disaster recovery problem in the research area of anomaly scene
understanding. Finally, we discuss future works to improve more robust,
explainable applications for effective initial responses.
| [
{
"version": "v1",
"created": "Mon, 5 Jun 2023 00:44:39 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 08:38:29 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Yasuno",
"Takato",
""
],
[
"Okano",
"Masahiro",
""
],
[
"Fujii",
"Junichiro",
""
]
] | TITLE: Disaster Anomaly Detector via Deeper FCDDs for Explainable Initial
Responses
ABSTRACT: Extreme natural disasters can have devastating effects on both urban and
rural areas. In any disaster event, an initial response is the key to rescue
within 72 hours and prompt recovery. During the initial stage of disaster
response, it is important to quickly assess the damage over a wide area and
identify priority areas. Among machine learning algorithms, deep anomaly
detection is effective in detecting devastation features that are different
from everyday features. In addition, explainable computer vision applications
should justify the initial responses. In this paper, we propose an anomaly
detection application utilizing deeper fully convolutional data descriptions
(FCDDs), that enables the localization of devastation features and
visualization of damage-marked heatmaps. More specifically, we show numerous
training and test results for a dataset AIDER with the four disaster
categories: collapsed buildings, traffic incidents, fires, and flooded areas.
We also implement ablation studies of anomalous class imbalance and the data
scale competing against the normal class. Our experiments provide results of
high accuracies over 95% for F1. Furthermore, we found that the deeper FCDD
with a VGG16 backbone consistently outperformed other baselines CNN27,
ResNet101, and Inceptionv3. This study presents a new solution that offers a
disaster anomaly detection application for initial responses with higher
accuracy and devastation explainability, providing a novel contribution to the
prompt disaster recovery problem in the research area of anomaly scene
understanding. Finally, we discuss future works to improve more robust,
explainable applications for effective initial responses.
|
2306.02548 | Yuhao Huang | Xinrui Zhou, Yuhao Huang, Wufeng Xue, Xin Yang, Yuxin Zou, Qilong
Ying, Yuanji Zhang, Jia Liu, Jie Ren, Dong Ni | Inflated 3D Convolution-Transformer for Weakly-supervised Carotid
Stenosis Grading with Ultrasound Videos | Accepted by MICCAI 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Localization of the narrowest position of the vessel and corresponding vessel
and remnant vessel delineation in carotid ultrasound (US) are essential for
carotid stenosis grading (CSG) in clinical practice. However, the pipeline is
time-consuming and tough due to the ambiguous boundaries of plaque and temporal
variation. To automatize this procedure, a large number of manual delineations
are usually required, which is not only laborious but also not reliable given
the annotation difficulty. In this study, we present the first video
classification framework for automatic CSG. Our contribution is three-fold.
First, to avoid the requirement of laborious and unreliable annotation, we
propose a novel and effective video classification network for
weakly-supervised CSG. Second, to ease the model training, we adopt an
inflation strategy for the network, where pre-trained 2D convolution weights
can be adapted into the 3D counterpart in our network for an effective warm
start. Third, to enhance the feature discrimination of the video, we propose a
novel attention-guided multi-dimension fusion (AMDF) transformer encoder to
model and integrate global dependencies within and across spatial and temporal
dimensions, where two lightweight cross-dimensional attention mechanisms are
designed. Our approach is extensively validated on a large clinically collected
carotid US video dataset, demonstrating state-of-the-art performance compared
with strong competitors.
| [
{
"version": "v1",
"created": "Mon, 5 Jun 2023 02:50:06 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 08:55:47 GMT"
},
{
"version": "v3",
"created": "Mon, 12 Jun 2023 09:28:07 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zhou",
"Xinrui",
""
],
[
"Huang",
"Yuhao",
""
],
[
"Xue",
"Wufeng",
""
],
[
"Yang",
"Xin",
""
],
[
"Zou",
"Yuxin",
""
],
[
"Ying",
"Qilong",
""
],
[
"Zhang",
"Yuanji",
""
],
[
"Liu",
"Jia",
""
],
[
"Ren",
"Jie",
""
],
[
"Ni",
"Dong",
""
]
] | TITLE: Inflated 3D Convolution-Transformer for Weakly-supervised Carotid
Stenosis Grading with Ultrasound Videos
ABSTRACT: Localization of the narrowest position of the vessel and corresponding vessel
and remnant vessel delineation in carotid ultrasound (US) are essential for
carotid stenosis grading (CSG) in clinical practice. However, the pipeline is
time-consuming and tough due to the ambiguous boundaries of plaque and temporal
variation. To automatize this procedure, a large number of manual delineations
are usually required, which is not only laborious but also not reliable given
the annotation difficulty. In this study, we present the first video
classification framework for automatic CSG. Our contribution is three-fold.
First, to avoid the requirement of laborious and unreliable annotation, we
propose a novel and effective video classification network for
weakly-supervised CSG. Second, to ease the model training, we adopt an
inflation strategy for the network, where pre-trained 2D convolution weights
can be adapted into the 3D counterpart in our network for an effective warm
start. Third, to enhance the feature discrimination of the video, we propose a
novel attention-guided multi-dimension fusion (AMDF) transformer encoder to
model and integrate global dependencies within and across spatial and temporal
dimensions, where two lightweight cross-dimensional attention mechanisms are
designed. Our approach is extensively validated on a large clinically collected
carotid US video dataset, demonstrating state-of-the-art performance compared
with strong competitors.
|
2306.03503 | Jose Berengueres Ph.D | Jose Berengueres and Marybeth Sandell | Applying Standards to Advance Upstream & Downstream Ethics in Large
Language Models | 8 pages, 4 tables, 2 figures | null | null | null | cs.CY cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper explores how AI-owners can develop safeguards for AI-generated
content by drawing from established codes of conduct and ethical standards in
other content-creation industries. It delves into the current state of ethical
awareness on Large Language Models (LLMs). By dissecting the mechanism of
content generation by LLMs, four key areas (upstream/downstream and at user
prompt/answer), where safeguards could be effectively applied, are identified.
A comparative analysis of these four areas follows and includes an evaluation
of the existing ethical safeguards in terms of cost, effectiveness, and
alignment with established industry practices. The paper's key argument is that
existing IT-related ethical codes, while adequate for traditional IT
engineering, are inadequate for the challenges posed by LLM-based content
generation. Drawing from established practices within journalism, we propose
potential standards for businesses involved in distributing and selling
LLM-generated content. Finally, potential conflicts of interest between dataset
curation at upstream and ethical benchmarking downstream are highlighted to
underscore the need for a broader evaluation beyond mere output. This study
prompts a nuanced conversation around ethical implications in this rapidly
evolving field of content generation.
| [
{
"version": "v1",
"created": "Tue, 6 Jun 2023 08:47:42 GMT"
},
{
"version": "v2",
"created": "Sun, 11 Jun 2023 10:45:57 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Berengueres",
"Jose",
""
],
[
"Sandell",
"Marybeth",
""
]
] | TITLE: Applying Standards to Advance Upstream & Downstream Ethics in Large
Language Models
ABSTRACT: This paper explores how AI-owners can develop safeguards for AI-generated
content by drawing from established codes of conduct and ethical standards in
other content-creation industries. It delves into the current state of ethical
awareness on Large Language Models (LLMs). By dissecting the mechanism of
content generation by LLMs, four key areas (upstream/downstream and at user
prompt/answer), where safeguards could be effectively applied, are identified.
A comparative analysis of these four areas follows and includes an evaluation
of the existing ethical safeguards in terms of cost, effectiveness, and
alignment with established industry practices. The paper's key argument is that
existing IT-related ethical codes, while adequate for traditional IT
engineering, are inadequate for the challenges posed by LLM-based content
generation. Drawing from established practices within journalism, we propose
potential standards for businesses involved in distributing and selling
LLM-generated content. Finally, potential conflicts of interest between dataset
curation at upstream and ethical benchmarking downstream are highlighted to
underscore the need for a broader evaluation beyond mere output. This study
prompts a nuanced conversation around ethical implications in this rapidly
evolving field of content generation.
|
2306.03974 | Hao Fei | Jiang Liu, Hao Fei, Fei Li, Jingye Li, Bobo Li, Liang Zhao, Chong Teng
and Donghong Ji | TKDP: Threefold Knowledge-enriched Deep Prompt Tuning for Few-shot Named
Entity Recognition | under review | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Few-shot named entity recognition (NER) exploits limited annotated instances
to identify named mentions. Effectively transferring the internal or external
resources thus becomes the key to few-shot NER. While the existing prompt
tuning methods have shown remarkable few-shot performances, they still fail to
make full use of knowledge. In this work, we investigate the integration of
rich knowledge to prompt tuning for stronger few-shot NER. We propose
incorporating the deep prompt tuning framework with threefold knowledge (namely
TKDP), including the internal 1) context knowledge and the external 2) label
knowledge & 3) sememe knowledge. TKDP encodes the three feature sources and
incorporates them into the soft prompt embeddings, which are further injected
into an existing pre-trained language model to facilitate predictions. On five
benchmark datasets, our knowledge-enriched model boosts by at most 11.53% F1
over the raw deep prompt method, and significantly outperforms 8
strong-performing baseline systems in 5-/10-/20-shot settings, showing great
potential in few-shot NER. Our TKDP can be broadly adapted to other few-shot
tasks without effort.
| [
{
"version": "v1",
"created": "Tue, 6 Jun 2023 19:11:59 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 06:50:35 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Liu",
"Jiang",
""
],
[
"Fei",
"Hao",
""
],
[
"Li",
"Fei",
""
],
[
"Li",
"Jingye",
""
],
[
"Li",
"Bobo",
""
],
[
"Zhao",
"Liang",
""
],
[
"Teng",
"Chong",
""
],
[
"Ji",
"Donghong",
""
]
] | TITLE: TKDP: Threefold Knowledge-enriched Deep Prompt Tuning for Few-shot Named
Entity Recognition
ABSTRACT: Few-shot named entity recognition (NER) exploits limited annotated instances
to identify named mentions. Effectively transferring the internal or external
resources thus becomes the key to few-shot NER. While the existing prompt
tuning methods have shown remarkable few-shot performances, they still fail to
make full use of knowledge. In this work, we investigate the integration of
rich knowledge to prompt tuning for stronger few-shot NER. We propose
incorporating the deep prompt tuning framework with threefold knowledge (namely
TKDP), including the internal 1) context knowledge and the external 2) label
knowledge & 3) sememe knowledge. TKDP encodes the three feature sources and
incorporates them into the soft prompt embeddings, which are further injected
into an existing pre-trained language model to facilitate predictions. On five
benchmark datasets, our knowledge-enriched model boosts by at most 11.53% F1
over the raw deep prompt method, and significantly outperforms 8
strong-performing baseline systems in 5-/10-/20-shot settings, showing great
potential in few-shot NER. Our TKDP can be broadly adapted to other few-shot
tasks without effort.
|
2306.03975 | Hao Fei | Bobo Li, Hao Fei, Fei Li, Shengqiong Wu, Lizi Liao, Yinwei Wei,
Tat-Seng Chua, Donghong Ji | Revisiting Conversation Discourse for Dialogue Disentanglement | under review | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Dialogue disentanglement aims to detach the chronologically ordered
utterances into several independent sessions. Conversation utterances are
essentially organized and described by the underlying discourse, and thus
dialogue disentanglement requires the full understanding and harnessing of the
intrinsic discourse attribute. In this paper, we propose enhancing dialogue
disentanglement by taking full advantage of the dialogue discourse
characteristics. First of all, in feature encoding stage, we construct the
heterogeneous graph representations to model the various dialogue-specific
discourse structural features, including the static speaker-role structures
(i.e., speaker-utterance and speaker-mentioning structure) and the dynamic
contextual structures (i.e., the utterance-distance and partial-replying
structure). We then develop a structure-aware framework to integrate the rich
structural features for better modeling the conversational semantic context.
Second, in model learning stage, we perform optimization with a hierarchical
ranking loss mechanism, which groups dialogue utterances into different
discourse levels and carries training covering pair-wise and session-wise
levels hierarchically. Third, in inference stage, we devise an easy-first
decoding algorithm, which performs utterance pairing under the easy-to-hard
manner with a global context, breaking the constraint of traditional sequential
decoding order. On two benchmark datasets, our overall system achieves new
state-of-the-art performances on all evaluations. In-depth analyses further
demonstrate the efficacy of each proposed idea and also reveal how our methods
help advance the task. Our work has great potential to facilitate broader
multi-party multi-thread dialogue applications.
| [
{
"version": "v1",
"created": "Tue, 6 Jun 2023 19:17:47 GMT"
},
{
"version": "v2",
"created": "Sat, 10 Jun 2023 06:42:03 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Li",
"Bobo",
""
],
[
"Fei",
"Hao",
""
],
[
"Li",
"Fei",
""
],
[
"Wu",
"Shengqiong",
""
],
[
"Liao",
"Lizi",
""
],
[
"Wei",
"Yinwei",
""
],
[
"Chua",
"Tat-Seng",
""
],
[
"Ji",
"Donghong",
""
]
] | TITLE: Revisiting Conversation Discourse for Dialogue Disentanglement
ABSTRACT: Dialogue disentanglement aims to detach the chronologically ordered
utterances into several independent sessions. Conversation utterances are
essentially organized and described by the underlying discourse, and thus
dialogue disentanglement requires the full understanding and harnessing of the
intrinsic discourse attribute. In this paper, we propose enhancing dialogue
disentanglement by taking full advantage of the dialogue discourse
characteristics. First of all, in feature encoding stage, we construct the
heterogeneous graph representations to model the various dialogue-specific
discourse structural features, including the static speaker-role structures
(i.e., speaker-utterance and speaker-mentioning structure) and the dynamic
contextual structures (i.e., the utterance-distance and partial-replying
structure). We then develop a structure-aware framework to integrate the rich
structural features for better modeling the conversational semantic context.
Second, in model learning stage, we perform optimization with a hierarchical
ranking loss mechanism, which groups dialogue utterances into different
discourse levels and carries training covering pair-wise and session-wise
levels hierarchically. Third, in inference stage, we devise an easy-first
decoding algorithm, which performs utterance pairing under the easy-to-hard
manner with a global context, breaking the constraint of traditional sequential
decoding order. On two benchmark datasets, our overall system achieves new
state-of-the-art performances on all evaluations. In-depth analyses further
demonstrate the efficacy of each proposed idea and also reveal how our methods
help advance the task. Our work has great potential to facilitate broader
multi-party multi-thread dialogue applications.
|
2306.04717 | Chunyi Li | Chunyi Li, Zicheng Zhang, Haoning Wu, Wei Sun, Xiongkuo Min, Xiaohong
Liu, Guangtao Zhai, Weisi Lin | AGIQA-3K: An Open Database for AI-Generated Image Quality Assessment | 12 pages, 11 figures | null | null | null | cs.CV cs.AI eess.IV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | With the rapid advancements of the text-to-image generative model,
AI-generated images (AGIs) have been widely applied to entertainment,
education, social media, etc. However, considering the large quality variance
among different AGIs, there is an urgent need for quality models that are
consistent with human subjective ratings. To address this issue, we extensively
consider various popular AGI models, generated AGI through different prompts
and model parameters, and collected subjective scores at the perceptual quality
and text-to-image alignment, thus building the most comprehensive AGI
subjective quality database AGIQA-3K so far. Furthermore, we conduct a
benchmark experiment on this database to evaluate the consistency between the
current Image Quality Assessment (IQA) model and human perception, while
proposing StairReward that significantly improves the assessment performance of
subjective text-to-image alignment. We believe that the fine-grained subjective
scores in AGIQA-3K will inspire subsequent AGI quality models to fit human
subjective perception mechanisms at both perception and alignment levels and to
optimize the generation result of future AGI models. The database is released
on https://github.com/lcysyzxdxc/AGIQA-3k-Database.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 18:28:21 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 16:42:59 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Li",
"Chunyi",
""
],
[
"Zhang",
"Zicheng",
""
],
[
"Wu",
"Haoning",
""
],
[
"Sun",
"Wei",
""
],
[
"Min",
"Xiongkuo",
""
],
[
"Liu",
"Xiaohong",
""
],
[
"Zhai",
"Guangtao",
""
],
[
"Lin",
"Weisi",
""
]
] | TITLE: AGIQA-3K: An Open Database for AI-Generated Image Quality Assessment
ABSTRACT: With the rapid advancements of the text-to-image generative model,
AI-generated images (AGIs) have been widely applied to entertainment,
education, social media, etc. However, considering the large quality variance
among different AGIs, there is an urgent need for quality models that are
consistent with human subjective ratings. To address this issue, we extensively
consider various popular AGI models, generated AGI through different prompts
and model parameters, and collected subjective scores at the perceptual quality
and text-to-image alignment, thus building the most comprehensive AGI
subjective quality database AGIQA-3K so far. Furthermore, we conduct a
benchmark experiment on this database to evaluate the consistency between the
current Image Quality Assessment (IQA) model and human perception, while
proposing StairReward that significantly improves the assessment performance of
subjective text-to-image alignment. We believe that the fine-grained subjective
scores in AGIQA-3K will inspire subsequent AGI quality models to fit human
subjective perception mechanisms at both perception and alignment levels and to
optimize the generation result of future AGI models. The database is released
on https://github.com/lcysyzxdxc/AGIQA-3k-Database.
|
2306.05044 | Jonathan Dupuy | Jonathan Dupuy and Anis Benyoub | Sampling Visible GGX Normals with Spherical Caps | 6 pages, 5 figures | null | null | null | cs.GR | http://creativecommons.org/licenses/by/4.0/ | Importance sampling the distribution of visible GGX normals requires sampling
those of a hemisphere. In this work, we introduce a novel method for sampling
such visible normals. Our method builds upon the insight that a hemispherical
mirror reflects parallel light rays uniformly within a solid angle shaped as a
spherical cap. This spherical cap has the same apex as the hemispherical
mirror, and its aperture given by the angle formed by the orientation of that
apex and the direction of incident light rays. Based on this insight, we sample
GGX visible normals as halfway vectors between a given incident direction and
directions drawn from its associated spherical cap. Our resulting
implementation is even simpler than that of Heitz and leads to up to systematic
speed-ups in our benchmarks.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 08:54:27 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 15:36:04 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Dupuy",
"Jonathan",
""
],
[
"Benyoub",
"Anis",
""
]
] | TITLE: Sampling Visible GGX Normals with Spherical Caps
ABSTRACT: Importance sampling the distribution of visible GGX normals requires sampling
those of a hemisphere. In this work, we introduce a novel method for sampling
such visible normals. Our method builds upon the insight that a hemispherical
mirror reflects parallel light rays uniformly within a solid angle shaped as a
spherical cap. This spherical cap has the same apex as the hemispherical
mirror, and its aperture given by the angle formed by the orientation of that
apex and the direction of incident light rays. Based on this insight, we sample
GGX visible normals as halfway vectors between a given incident direction and
directions drawn from its associated spherical cap. Our resulting
implementation is even simpler than that of Heitz and leads to up to systematic
speed-ups in our benchmarks.
|
2306.05749 | Jiaxin Zhang | Jiaxin Zhang, Bangdong Chen, Hiuyi Cheng, Fengjun Guo, Kai Ding,
Lianwen Jin | DocAligner: Annotating Real-world Photographic Document Images by Simply
Taking Pictures | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Recently, there has been a growing interest in research concerning document
image analysis and recognition in photographic scenarios. However, the lack of
labeled datasets for this emerging challenge poses a significant obstacle, as
manual annotation can be time-consuming and impractical. To tackle this issue,
we present DocAligner, a novel method that streamlines the manual annotation
process to a simple step of taking pictures. DocAligner achieves this by
establishing dense correspondence between photographic document images and
their clean counterparts. It enables the automatic transfer of existing
annotations in clean document images to photographic ones and helps to
automatically acquire labels that are unavailable through manual labeling.
Considering the distinctive characteristics of document images, DocAligner
incorporates several innovative features. First, we propose a non-rigid
pre-alignment technique based on the document's edges, which effectively
eliminates interference caused by significant global shifts and repetitive
patterns present in document images. Second, to handle large shifts and ensure
high accuracy, we introduce a hierarchical aligning approach that combines
global and local correlation layers. Furthermore, considering the importance of
fine-grained elements in document images, we present a details recurrent
refinement module to enhance the output in a high-resolution space. To train
DocAligner, we construct a synthetic dataset and introduce a self-supervised
learning approach to enhance its robustness for real-world data. Through
extensive experiments, we demonstrate the effectiveness of DocAligner and the
acquired dataset. Datasets and codes will be publicly available.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 08:29:15 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 04:03:23 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zhang",
"Jiaxin",
""
],
[
"Chen",
"Bangdong",
""
],
[
"Cheng",
"Hiuyi",
""
],
[
"Guo",
"Fengjun",
""
],
[
"Ding",
"Kai",
""
],
[
"Jin",
"Lianwen",
""
]
] | TITLE: DocAligner: Annotating Real-world Photographic Document Images by Simply
Taking Pictures
ABSTRACT: Recently, there has been a growing interest in research concerning document
image analysis and recognition in photographic scenarios. However, the lack of
labeled datasets for this emerging challenge poses a significant obstacle, as
manual annotation can be time-consuming and impractical. To tackle this issue,
we present DocAligner, a novel method that streamlines the manual annotation
process to a simple step of taking pictures. DocAligner achieves this by
establishing dense correspondence between photographic document images and
their clean counterparts. It enables the automatic transfer of existing
annotations in clean document images to photographic ones and helps to
automatically acquire labels that are unavailable through manual labeling.
Considering the distinctive characteristics of document images, DocAligner
incorporates several innovative features. First, we propose a non-rigid
pre-alignment technique based on the document's edges, which effectively
eliminates interference caused by significant global shifts and repetitive
patterns present in document images. Second, to handle large shifts and ensure
high accuracy, we introduce a hierarchical aligning approach that combines
global and local correlation layers. Furthermore, considering the importance of
fine-grained elements in document images, we present a details recurrent
refinement module to enhance the output in a high-resolution space. To train
DocAligner, we construct a synthetic dataset and introduce a self-supervised
learning approach to enhance its robustness for real-world data. Through
extensive experiments, we demonstrate the effectiveness of DocAligner and the
acquired dataset. Datasets and codes will be publicly available.
|
2306.05775 | Zhengqing Miao | Zhengqing Miao and Meirong Zhao | Weight Freezing: A Regularization Approach for Fully Connected Layers
with an Application in EEG Classification | 16 pages, 5 figures | null | null | null | cs.LG eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the realm of EEG decoding, enhancing the performance of artificial neural
networks (ANNs) carries significant potential. This study introduces a novel
approach, termed "weight freezing", that is anchored on the principles of ANN
regularization and neuroscience prior knowledge. The concept of weight freezing
revolves around the idea of reducing certain neurons' influence on the
decision-making process for a specific EEG task by freezing specific weights in
the fully connected layer during the backpropagation process. This is
actualized through the use of a mask matrix and a threshold to determine the
proportion of weights to be frozen during backpropagation. Moreover, by setting
the masked weights to zero, weight freezing can not only realize sparse
connections in networks with a fully connected layer as the classifier but also
function as an efficacious regularization method for fully connected layers.
Through experiments involving three distinct ANN architectures and three widely
recognized EEG datasets, we validate the potency of weight freezing. Our method
significantly surpasses previous peak performances in classification accuracy
across all examined datasets. Supplementary control experiments offer insights
into performance differences pre and post weight freezing implementation and
scrutinize the influence of the threshold in the weight freezing process. Our
study underscores the superior efficacy of weight freezing compared to
traditional fully connected networks for EEG feature classification tasks. With
its proven effectiveness, this innovative approach holds substantial promise
for contributing to future strides in EEG decoding research.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 09:33:34 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 00:42:03 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Miao",
"Zhengqing",
""
],
[
"Zhao",
"Meirong",
""
]
] | TITLE: Weight Freezing: A Regularization Approach for Fully Connected Layers
with an Application in EEG Classification
ABSTRACT: In the realm of EEG decoding, enhancing the performance of artificial neural
networks (ANNs) carries significant potential. This study introduces a novel
approach, termed "weight freezing", that is anchored on the principles of ANN
regularization and neuroscience prior knowledge. The concept of weight freezing
revolves around the idea of reducing certain neurons' influence on the
decision-making process for a specific EEG task by freezing specific weights in
the fully connected layer during the backpropagation process. This is
actualized through the use of a mask matrix and a threshold to determine the
proportion of weights to be frozen during backpropagation. Moreover, by setting
the masked weights to zero, weight freezing can not only realize sparse
connections in networks with a fully connected layer as the classifier but also
function as an efficacious regularization method for fully connected layers.
Through experiments involving three distinct ANN architectures and three widely
recognized EEG datasets, we validate the potency of weight freezing. Our method
significantly surpasses previous peak performances in classification accuracy
across all examined datasets. Supplementary control experiments offer insights
into performance differences pre and post weight freezing implementation and
scrutinize the influence of the threshold in the weight freezing process. Our
study underscores the superior efficacy of weight freezing compared to
traditional fully connected networks for EEG feature classification tasks. With
its proven effectiveness, this innovative approach holds substantial promise
for contributing to future strides in EEG decoding research.
|
2306.06062 | Oluwadamilola Fasina | Oluwadamilola Fasina, Guillaume Huguet, Alexander Tong, Yanlei Zhang,
Guy Wolf, Maximilian Nickel, Ian Adelstein, Smita Krishnaswamy | Neural FIM for learning Fisher Information Metrics from point cloud data | 13 pages, 11 figures, 1 table | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although data diffusion embeddings are ubiquitous in unsupervised learning
and have proven to be a viable technique for uncovering the underlying
intrinsic geometry of data, diffusion embeddings are inherently limited due to
their discrete nature. To this end, we propose neural FIM, a method for
computing the Fisher information metric (FIM) from point cloud data - allowing
for a continuous manifold model for the data. Neural FIM creates an extensible
metric space from discrete point cloud data such that information from the
metric can inform us of manifold characteristics such as volume and geodesics.
We demonstrate Neural FIM's utility in selecting parameters for the PHATE
visualization method as well as its ability to obtain information pertaining to
local volume illuminating branching points and cluster centers embeddings of a
toy dataset and two single-cell datasets of IPSC reprogramming and PBMCs
(immune cells).
| [
{
"version": "v1",
"created": "Thu, 1 Jun 2023 17:36:13 GMT"
},
{
"version": "v2",
"created": "Mon, 12 Jun 2023 00:37:58 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Fasina",
"Oluwadamilola",
""
],
[
"Huguet",
"Guillaume",
""
],
[
"Tong",
"Alexander",
""
],
[
"Zhang",
"Yanlei",
""
],
[
"Wolf",
"Guy",
""
],
[
"Nickel",
"Maximilian",
""
],
[
"Adelstein",
"Ian",
""
],
[
"Krishnaswamy",
"Smita",
""
]
] | TITLE: Neural FIM for learning Fisher Information Metrics from point cloud data
ABSTRACT: Although data diffusion embeddings are ubiquitous in unsupervised learning
and have proven to be a viable technique for uncovering the underlying
intrinsic geometry of data, diffusion embeddings are inherently limited due to
their discrete nature. To this end, we propose neural FIM, a method for
computing the Fisher information metric (FIM) from point cloud data - allowing
for a continuous manifold model for the data. Neural FIM creates an extensible
metric space from discrete point cloud data such that information from the
metric can inform us of manifold characteristics such as volume and geodesics.
We demonstrate Neural FIM's utility in selecting parameters for the PHATE
visualization method as well as its ability to obtain information pertaining to
local volume illuminating branching points and cluster centers embeddings of a
toy dataset and two single-cell datasets of IPSC reprogramming and PBMCs
(immune cells).
|
2306.06107 | Paul Stahlhofen | Paul Stahlhofen, Andr\'e Artelt, Luca Hermes, Barbara Hammer | Adversarial Attacks on Leakage Detectors in Water Distribution Networks | This paper was accepted for the 17th International Work-Conference on
Artificial Neural Networks (IWANN 2023). A link to the version of record will
be provided upon publication of the conference proceedings | null | null | null | cs.CR cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | Many Machine Learning models are vulnerable to adversarial attacks: There
exist methodologies that add a small (imperceptible) perturbation to an input
such that the model comes up with a wrong prediction. Better understanding of
such attacks is crucial in particular for models used in security-critical
domains, such as monitoring of water distribution networks, in order to devise
counter-measures enhancing model robustness and trustworthiness.
We propose a taxonomy for adversarial attacks against machine learning based
leakage detectors in water distribution networks. Following up on this, we
focus on a particular type of attack: an adversary searching the least
sensitive point, that is, the location in the water network where the largest
possible undetected leak could occur. Based on a mathematical formalization of
the least sensitive point problem, we use three different algorithmic
approaches to find a solution. Results are evaluated on two benchmark water
distribution networks.
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 12:05:18 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Stahlhofen",
"Paul",
""
],
[
"Artelt",
"André",
""
],
[
"Hermes",
"Luca",
""
],
[
"Hammer",
"Barbara",
""
]
] | TITLE: Adversarial Attacks on Leakage Detectors in Water Distribution Networks
ABSTRACT: Many Machine Learning models are vulnerable to adversarial attacks: There
exist methodologies that add a small (imperceptible) perturbation to an input
such that the model comes up with a wrong prediction. Better understanding of
such attacks is crucial in particular for models used in security-critical
domains, such as monitoring of water distribution networks, in order to devise
counter-measures enhancing model robustness and trustworthiness.
We propose a taxonomy for adversarial attacks against machine learning based
leakage detectors in water distribution networks. Following up on this, we
focus on a particular type of attack: an adversary searching the least
sensitive point, that is, the location in the water network where the largest
possible undetected leak could occur. Based on a mathematical formalization of
the least sensitive point problem, we use three different algorithmic
approaches to find a solution. Results are evaluated on two benchmark water
distribution networks.
|
2306.06108 | Youssef Elmougy | Youssef Elmougy and Ling Liu | Demystifying Fraudulent Transactions and Illicit Nodes in the Bitcoin
Network for Financial Forensics | null | null | 10.1145/3580305.3599803 | null | cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Blockchain provides the unique and accountable channel for financial
forensics by mining its open and immutable transaction data. A recent surge has
been witnessed by training machine learning models with cryptocurrency
transaction data for anomaly detection, such as money laundering and other
fraudulent activities. This paper presents a holistic applied data science
approach to fraud detection in the Bitcoin network with two original
contributions. First, we contribute the Elliptic++ dataset, which extends the
Elliptic transaction dataset to include over 822k Bitcoin wallet addresses
(nodes), each with 56 features, and 1.27M temporal interactions. This enables
both the detection of fraudulent transactions and the detection of illicit
addresses (actors) in the Bitcoin network by leveraging four types of graph
data: (i) the transaction-to-transaction graph, representing the money flow in
the Bitcoin network, (ii) the address-to-address interaction graph, capturing
the types of transaction flows between Bitcoin addresses, (iii) the
address-transaction graph, representing the bi-directional money flow between
addresses and transactions (BTC flow from input address to one or more
transactions and BTC flow from a transaction to one or more output addresses),
and (iv) the user entity graph, capturing clusters of Bitcoin addresses
representing unique Bitcoin users. Second, we perform fraud detection tasks on
all four graphs by using diverse machine learning algorithms. We show that
adding enhanced features from the address-to-address and the
address-transaction graphs not only assists in effectively detecting both
illicit transactions and illicit addresses, but also assists in gaining
in-depth understanding of the root cause of money laundering vulnerabilities in
cryptocurrency transactions and the strategies for fraud detection and
prevention. Released at github.com/git-disl/EllipticPlusPlus.
| [
{
"version": "v1",
"created": "Thu, 25 May 2023 18:36:54 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Elmougy",
"Youssef",
""
],
[
"Liu",
"Ling",
""
]
] | TITLE: Demystifying Fraudulent Transactions and Illicit Nodes in the Bitcoin
Network for Financial Forensics
ABSTRACT: Blockchain provides the unique and accountable channel for financial
forensics by mining its open and immutable transaction data. A recent surge has
been witnessed by training machine learning models with cryptocurrency
transaction data for anomaly detection, such as money laundering and other
fraudulent activities. This paper presents a holistic applied data science
approach to fraud detection in the Bitcoin network with two original
contributions. First, we contribute the Elliptic++ dataset, which extends the
Elliptic transaction dataset to include over 822k Bitcoin wallet addresses
(nodes), each with 56 features, and 1.27M temporal interactions. This enables
both the detection of fraudulent transactions and the detection of illicit
addresses (actors) in the Bitcoin network by leveraging four types of graph
data: (i) the transaction-to-transaction graph, representing the money flow in
the Bitcoin network, (ii) the address-to-address interaction graph, capturing
the types of transaction flows between Bitcoin addresses, (iii) the
address-transaction graph, representing the bi-directional money flow between
addresses and transactions (BTC flow from input address to one or more
transactions and BTC flow from a transaction to one or more output addresses),
and (iv) the user entity graph, capturing clusters of Bitcoin addresses
representing unique Bitcoin users. Second, we perform fraud detection tasks on
all four graphs by using diverse machine learning algorithms. We show that
adding enhanced features from the address-to-address and the
address-transaction graphs not only assists in effectively detecting both
illicit transactions and illicit addresses, but also assists in gaining
in-depth understanding of the root cause of money laundering vulnerabilities in
cryptocurrency transactions and the strategies for fraud detection and
prevention. Released at github.com/git-disl/EllipticPlusPlus.
|
2306.06109 | Van Nguyen | Michael Fu, Trung Le, Van Nguyen, Chakkrit Tantithamthavorn, Dinh
Phung | Learning to Quantize Vulnerability Patterns and Match to Locate
Statement-Level Vulnerabilities | null | null | null | null | cs.CR cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Deep learning (DL) models have become increasingly popular in identifying
software vulnerabilities. Prior studies found that vulnerabilities across
different vulnerable programs may exhibit similar vulnerable scopes, implicitly
forming discernible vulnerability patterns that can be learned by DL models
through supervised training. However, vulnerable scopes still manifest in
various spatial locations and formats within a program, posing challenges for
models to accurately identify vulnerable statements. Despite this challenge,
state-of-the-art vulnerability detection approaches fail to exploit the
vulnerability patterns that arise in vulnerable programs. To take full
advantage of vulnerability patterns and unleash the ability of DL models, we
propose a novel vulnerability-matching approach in this paper, drawing
inspiration from program analysis tools that locate vulnerabilities based on
pre-defined patterns. Specifically, a vulnerability codebook is learned, which
consists of quantized vectors representing various vulnerability patterns.
During inference, the codebook is iterated to match all learned patterns and
predict the presence of potential vulnerabilities within a given program. Our
approach was extensively evaluated on a real-world dataset comprising more than
188,000 C/C++ functions. The evaluation results show that our approach achieves
an F1-score of 94% (6% higher than the previous best) and 82% (19% higher than
the previous best) for function and statement-level vulnerability
identification, respectively. These substantial enhancements highlight the
effectiveness of our approach to identifying vulnerabilities. The training code
and pre-trained models are available at https://github.com/optimatch/optimatch.
| [
{
"version": "v1",
"created": "Fri, 26 May 2023 04:13:31 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Fu",
"Michael",
""
],
[
"Le",
"Trung",
""
],
[
"Nguyen",
"Van",
""
],
[
"Tantithamthavorn",
"Chakkrit",
""
],
[
"Phung",
"Dinh",
""
]
] | TITLE: Learning to Quantize Vulnerability Patterns and Match to Locate
Statement-Level Vulnerabilities
ABSTRACT: Deep learning (DL) models have become increasingly popular in identifying
software vulnerabilities. Prior studies found that vulnerabilities across
different vulnerable programs may exhibit similar vulnerable scopes, implicitly
forming discernible vulnerability patterns that can be learned by DL models
through supervised training. However, vulnerable scopes still manifest in
various spatial locations and formats within a program, posing challenges for
models to accurately identify vulnerable statements. Despite this challenge,
state-of-the-art vulnerability detection approaches fail to exploit the
vulnerability patterns that arise in vulnerable programs. To take full
advantage of vulnerability patterns and unleash the ability of DL models, we
propose a novel vulnerability-matching approach in this paper, drawing
inspiration from program analysis tools that locate vulnerabilities based on
pre-defined patterns. Specifically, a vulnerability codebook is learned, which
consists of quantized vectors representing various vulnerability patterns.
During inference, the codebook is iterated to match all learned patterns and
predict the presence of potential vulnerabilities within a given program. Our
approach was extensively evaluated on a real-world dataset comprising more than
188,000 C/C++ functions. The evaluation results show that our approach achieves
an F1-score of 94% (6% higher than the previous best) and 82% (19% higher than
the previous best) for function and statement-level vulnerability
identification, respectively. These substantial enhancements highlight the
effectiveness of our approach to identifying vulnerabilities. The training code
and pre-trained models are available at https://github.com/optimatch/optimatch.
|
2306.06110 | Binyang Song | Binyang Song, Chenyang Yuan, Frank Permenter, Nikos Arechiga, Faez
Ahmed | Surrogate Modeling of Car Drag Coefficient with Depth and Normal
Renderings | null | null | null | null | cs.LG cs.CE cs.CV | http://creativecommons.org/licenses/by/4.0/ | Generative AI models have made significant progress in automating the
creation of 3D shapes, which has the potential to transform car design. In
engineering design and optimization, evaluating engineering metrics is crucial.
To make generative models performance-aware and enable them to create
high-performing designs, surrogate modeling of these metrics is necessary.
However, the currently used representations of three-dimensional (3D) shapes
either require extensive computational resources to learn or suffer from
significant information loss, which impairs their effectiveness in surrogate
modeling. To address this issue, we propose a new two-dimensional (2D)
representation of 3D shapes. We develop a surrogate drag model based on this
representation to verify its effectiveness in predicting 3D car drag. We
construct a diverse dataset of 9,070 high-quality 3D car meshes labeled by drag
coefficients computed from computational fluid dynamics (CFD) simulations to
train our model. Our experiments demonstrate that our model can accurately and
efficiently evaluate drag coefficients with an $R^2$ value above 0.84 for
various car categories. Moreover, the proposed representation method can be
generalized to many other product categories beyond cars. Our model is
implemented using deep neural networks, making it compatible with recent AI
image generation tools (such as Stable Diffusion) and a significant step
towards the automatic generation of drag-optimized car designs. We have made
the dataset and code publicly available at
https://decode.mit.edu/projects/dragprediction/.
| [
{
"version": "v1",
"created": "Fri, 26 May 2023 09:33:12 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Song",
"Binyang",
""
],
[
"Yuan",
"Chenyang",
""
],
[
"Permenter",
"Frank",
""
],
[
"Arechiga",
"Nikos",
""
],
[
"Ahmed",
"Faez",
""
]
] | TITLE: Surrogate Modeling of Car Drag Coefficient with Depth and Normal
Renderings
ABSTRACT: Generative AI models have made significant progress in automating the
creation of 3D shapes, which has the potential to transform car design. In
engineering design and optimization, evaluating engineering metrics is crucial.
To make generative models performance-aware and enable them to create
high-performing designs, surrogate modeling of these metrics is necessary.
However, the currently used representations of three-dimensional (3D) shapes
either require extensive computational resources to learn or suffer from
significant information loss, which impairs their effectiveness in surrogate
modeling. To address this issue, we propose a new two-dimensional (2D)
representation of 3D shapes. We develop a surrogate drag model based on this
representation to verify its effectiveness in predicting 3D car drag. We
construct a diverse dataset of 9,070 high-quality 3D car meshes labeled by drag
coefficients computed from computational fluid dynamics (CFD) simulations to
train our model. Our experiments demonstrate that our model can accurately and
efficiently evaluate drag coefficients with an $R^2$ value above 0.84 for
various car categories. Moreover, the proposed representation method can be
generalized to many other product categories beyond cars. Our model is
implemented using deep neural networks, making it compatible with recent AI
image generation tools (such as Stable Diffusion) and a significant step
towards the automatic generation of drag-optimized car designs. We have made
the dataset and code publicly available at
https://decode.mit.edu/projects/dragprediction/.
|
2306.06113 | Xiao-Feng Zhang | Xiaofeng Zhang, Chaochen Gu, Shanying Zhu | SAM-helps-Shadow:When Segment Anything Model meet shadow removal | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The challenges surrounding the application of image shadow removal to
real-world images and not just constrained datasets like ISTD/SRD have
highlighted an urgent need for zero-shot learning in this field. In this study,
we innovatively adapted the SAM (Segment anything model) for shadow removal by
introducing SAM-helps-Shadow, effectively integrating shadow detection and
removal into a single stage. Our approach utilized the model's detection
results as a potent prior for facilitating shadow detection, followed by shadow
removal using a second-order deep unfolding network. The source code of
SAM-helps-Shadow can be obtained from
https://github.com/zhangbaijin/SAM-helps-Shadow.
| [
{
"version": "v1",
"created": "Thu, 1 Jun 2023 06:37:19 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Zhang",
"Xiaofeng",
""
],
[
"Gu",
"Chaochen",
""
],
[
"Zhu",
"Shanying",
""
]
] | TITLE: SAM-helps-Shadow:When Segment Anything Model meet shadow removal
ABSTRACT: The challenges surrounding the application of image shadow removal to
real-world images and not just constrained datasets like ISTD/SRD have
highlighted an urgent need for zero-shot learning in this field. In this study,
we innovatively adapted the SAM (Segment anything model) for shadow removal by
introducing SAM-helps-Shadow, effectively integrating shadow detection and
removal into a single stage. Our approach utilized the model's detection
results as a potent prior for facilitating shadow detection, followed by shadow
removal using a second-order deep unfolding network. The source code of
SAM-helps-Shadow can be obtained from
https://github.com/zhangbaijin/SAM-helps-Shadow.
|
2306.06117 | Shawan Mohammed | Shawan Mohammed, Hannah Siebers, Ted Preu{\ss} | Strengths and Weaknesses of 3D Pose Estimation and Inertial Motion
Capture System for Movement Therapy | null | null | null | null | eess.IV cs.AI cs.CV cs.LG eess.SP | http://creativecommons.org/licenses/by/4.0/ | 3D pose estimation offers the opportunity for fast, non-invasive, and
accurate motion analysis. This is of special interest also for clinical use.
Currently, motion capture systems are used, as they offer robust and precise
data acquisition, which is essential in the case of clinical applications. In
this study, we investigate the accuracy of the state-of-the-art 3D position
estimation approach MeTrabs, compared to the established inertial sensor system
MTw Awinda for specific motion exercises. The study uses and provides an
evaluation dataset of parallel recordings from 10 subjects during various
movement therapy exercises. The information from the Awinda system and the
frames for monocular pose estimation are synchronized. For the comparison,
clinically relevant parameters for joint angles of ankle, knee, back, and elbow
flexion-extension were estimated and evaluated using mean, median, and maximum
deviation between the calculated joint angles for the different exercises,
camera positions, and clothing items. The results of the analysis indicate that
the mean and median deviations can be kept below 5{\deg} for some of the
studied angles. These joints could be considered for medical applications even
considering the maximum deviations of 15{\deg}. However, caution should be
applied to certain particularly problematic joints. In particular, elbow
flexions, which showed high maximum deviations of up to 50{\deg} in our
analysis. Furthermore, the type of exercise plays a crucial role in the
reliable and safe application of the 3D position estimation method. For
example, all joint angles showed a significant deterioration in performance
during exercises near the ground.
| [
{
"version": "v1",
"created": "Thu, 1 Jun 2023 20:35:06 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Mohammed",
"Shawan",
""
],
[
"Siebers",
"Hannah",
""
],
[
"Preuß",
"Ted",
""
]
] | TITLE: Strengths and Weaknesses of 3D Pose Estimation and Inertial Motion
Capture System for Movement Therapy
ABSTRACT: 3D pose estimation offers the opportunity for fast, non-invasive, and
accurate motion analysis. This is of special interest also for clinical use.
Currently, motion capture systems are used, as they offer robust and precise
data acquisition, which is essential in the case of clinical applications. In
this study, we investigate the accuracy of the state-of-the-art 3D position
estimation approach MeTrabs, compared to the established inertial sensor system
MTw Awinda for specific motion exercises. The study uses and provides an
evaluation dataset of parallel recordings from 10 subjects during various
movement therapy exercises. The information from the Awinda system and the
frames for monocular pose estimation are synchronized. For the comparison,
clinically relevant parameters for joint angles of ankle, knee, back, and elbow
flexion-extension were estimated and evaluated using mean, median, and maximum
deviation between the calculated joint angles for the different exercises,
camera positions, and clothing items. The results of the analysis indicate that
the mean and median deviations can be kept below 5{\deg} for some of the
studied angles. These joints could be considered for medical applications even
considering the maximum deviations of 15{\deg}. However, caution should be
applied to certain particularly problematic joints. In particular, elbow
flexions, which showed high maximum deviations of up to 50{\deg} in our
analysis. Furthermore, the type of exercise plays a crucial role in the
reliable and safe application of the 3D position estimation method. For
example, all joint angles showed a significant deterioration in performance
during exercises near the ground.
|
2306.06129 | Matteo Risso | Alessio Burrello, Matteo Risso, Noemi Tomasello, Yukai Chen, Luca
Benini, Enrico Macii, Massimo Poncino, Daniele Jahier Pagliari | Energy-efficient Wearable-to-Mobile Offload of ML Inference for
PPG-based Heart-Rate Estimation | Published at 2023 Design, Automation \& Test in Europe Conference \&
Exhibition (DATE) | 2023 Design, Automation & Test in Europe Conference & Exhibition
(DATE) | 10.23919/DATE56975.2023.10137129 | null | eess.SP cs.LG | http://creativecommons.org/licenses/by/4.0/ | Modern smartwatches often include photoplethysmographic (PPG) sensors to
measure heartbeats or blood pressure through complex algorithms that fuse PPG
data with other signals. In this work, we propose a collaborative inference
approach that uses both a smartwatch and a connected smartphone to maximize the
performance of heart rate (HR) tracking while also maximizing the smartwatch's
battery life. In particular, we first analyze the trade-offs between running
on-device HR tracking or offloading the work to the mobile. Then, thanks to an
additional step to evaluate the difficulty of the upcoming HR prediction, we
demonstrate that we can smartly manage the workload between smartwatch and
smartphone, maintaining a low mean absolute error (MAE) while reducing energy
consumption. We benchmark our approach on a custom smartwatch prototype,
including the STM32WB55 MCU and Bluetooth Low-Energy (BLE) communication, and a
Raspberry Pi3 as a proxy for the smartphone. With our Collaborative Heart Rate
Inference System (CHRIS), we obtain a set of Pareto-optimal configurations
demonstrating the same MAE as State-of-Art (SoA) algorithms while consuming
less energy. For instance, we can achieve approximately the same MAE of
TimePPG-Small (5.54 BPM MAE vs. 5.60 BPM MAE) while reducing the energy by
2.03x, with a configuration that offloads 80\% of the predictions to the phone.
Furthermore, accepting a performance degradation to 7.16 BPM of MAE, we can
achieve an energy consumption of 179 uJ per prediction, 3.03x less than running
TimePPG-Small on the smartwatch, and 1.82x less than streaming all the input
data to the phone.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 10:12:50 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Burrello",
"Alessio",
""
],
[
"Risso",
"Matteo",
""
],
[
"Tomasello",
"Noemi",
""
],
[
"Chen",
"Yukai",
""
],
[
"Benini",
"Luca",
""
],
[
"Macii",
"Enrico",
""
],
[
"Poncino",
"Massimo",
""
],
[
"Pagliari",
"Daniele Jahier",
""
]
] | TITLE: Energy-efficient Wearable-to-Mobile Offload of ML Inference for
PPG-based Heart-Rate Estimation
ABSTRACT: Modern smartwatches often include photoplethysmographic (PPG) sensors to
measure heartbeats or blood pressure through complex algorithms that fuse PPG
data with other signals. In this work, we propose a collaborative inference
approach that uses both a smartwatch and a connected smartphone to maximize the
performance of heart rate (HR) tracking while also maximizing the smartwatch's
battery life. In particular, we first analyze the trade-offs between running
on-device HR tracking or offloading the work to the mobile. Then, thanks to an
additional step to evaluate the difficulty of the upcoming HR prediction, we
demonstrate that we can smartly manage the workload between smartwatch and
smartphone, maintaining a low mean absolute error (MAE) while reducing energy
consumption. We benchmark our approach on a custom smartwatch prototype,
including the STM32WB55 MCU and Bluetooth Low-Energy (BLE) communication, and a
Raspberry Pi3 as a proxy for the smartphone. With our Collaborative Heart Rate
Inference System (CHRIS), we obtain a set of Pareto-optimal configurations
demonstrating the same MAE as State-of-Art (SoA) algorithms while consuming
less energy. For instance, we can achieve approximately the same MAE of
TimePPG-Small (5.54 BPM MAE vs. 5.60 BPM MAE) while reducing the energy by
2.03x, with a configuration that offloads 80\% of the predictions to the phone.
Furthermore, accepting a performance degradation to 7.16 BPM of MAE, we can
achieve an energy consumption of 179 uJ per prediction, 3.03x less than running
TimePPG-Small on the smartwatch, and 1.82x less than streaming all the input
data to the phone.
|
2306.06130 | Gonzalo Mart\'inez | Gonzalo Mart\'inez, Lauren Watson, Pedro Reviriego, Jos\'e Alberto
Hern\'andez, Marc Juarez, Rik Sarkar | Towards Understanding the Interplay of Generative Artificial
Intelligence and the Internet | null | null | null | null | cs.AI cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid adoption of generative Artificial Intelligence (AI) tools that can
generate realistic images or text, such as DALL-E, MidJourney, or ChatGPT, have
put the societal impacts of these technologies at the center of public debate.
These tools are possible due to the massive amount of data (text and images)
that is publicly available through the Internet. At the same time, these
generative AI tools become content creators that are already contributing to
the data that is available to train future models. Therefore, future versions
of generative AI tools will be trained with a mix of human-created and
AI-generated content, causing a potential feedback loop between generative AI
and public data repositories. This interaction raises many questions: how will
future versions of generative AI tools behave when trained on a mixture of real
and AI generated data? Will they evolve and improve with the new data sets or
on the contrary will they degrade? Will evolution introduce biases or reduce
diversity in subsequent generations of generative AI tools? What are the
societal implications of the possible degradation of these models? Can we
mitigate the effects of this feedback loop? In this document, we explore the
effect of this interaction and report some initial results using simple
diffusion models trained with various image datasets. Our results show that the
quality and diversity of the generated images can degrade over time suggesting
that incorporating AI-created data can have undesired effects on future
versions of generative models.
| [
{
"version": "v1",
"created": "Thu, 8 Jun 2023 11:14:51 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Martínez",
"Gonzalo",
""
],
[
"Watson",
"Lauren",
""
],
[
"Reviriego",
"Pedro",
""
],
[
"Hernández",
"José Alberto",
""
],
[
"Juarez",
"Marc",
""
],
[
"Sarkar",
"Rik",
""
]
] | TITLE: Towards Understanding the Interplay of Generative Artificial
Intelligence and the Internet
ABSTRACT: The rapid adoption of generative Artificial Intelligence (AI) tools that can
generate realistic images or text, such as DALL-E, MidJourney, or ChatGPT, have
put the societal impacts of these technologies at the center of public debate.
These tools are possible due to the massive amount of data (text and images)
that is publicly available through the Internet. At the same time, these
generative AI tools become content creators that are already contributing to
the data that is available to train future models. Therefore, future versions
of generative AI tools will be trained with a mix of human-created and
AI-generated content, causing a potential feedback loop between generative AI
and public data repositories. This interaction raises many questions: how will
future versions of generative AI tools behave when trained on a mixture of real
and AI generated data? Will they evolve and improve with the new data sets or
on the contrary will they degrade? Will evolution introduce biases or reduce
diversity in subsequent generations of generative AI tools? What are the
societal implications of the possible degradation of these models? Can we
mitigate the effects of this feedback loop? In this document, we explore the
effect of this interaction and report some initial results using simple
diffusion models trained with various image datasets. Our results show that the
quality and diversity of the generated images can degrade over time suggesting
that incorporating AI-created data can have undesired effects on future
versions of generative models.
|
2306.06135 | Susan Hao | Susan Hao, Piyush Kumar, Sarah Laszlo, Shivani Poddar, Bhaktipriya
Radharapu, Renee Shelby | Safety and Fairness for Content Moderation in Generative Models | CVPR Workshop Paper | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | With significant advances in generative AI, new technologies are rapidly
being deployed with generative components. Generative models are typically
trained on large datasets, resulting in model behaviors that can mimic the
worst of the content in the training data. Responsible deployment of generative
technologies requires content moderation strategies, such as safety input and
output filters. Here, we provide a theoretical framework for conceptualizing
responsible content moderation of text-to-image generative technologies,
including a demonstration of how to empirically measure the constructs we
enumerate. We define and distinguish the concepts of safety, fairness, and
metric equity, and enumerate example harms that can come in each domain. We
then provide a demonstration of how the defined harms can be quantified. We
conclude with a summary of how the style of harms quantification we demonstrate
enables data-driven content moderation decisions.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 01:37:32 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Hao",
"Susan",
""
],
[
"Kumar",
"Piyush",
""
],
[
"Laszlo",
"Sarah",
""
],
[
"Poddar",
"Shivani",
""
],
[
"Radharapu",
"Bhaktipriya",
""
],
[
"Shelby",
"Renee",
""
]
] | TITLE: Safety and Fairness for Content Moderation in Generative Models
ABSTRACT: With significant advances in generative AI, new technologies are rapidly
being deployed with generative components. Generative models are typically
trained on large datasets, resulting in model behaviors that can mimic the
worst of the content in the training data. Responsible deployment of generative
technologies requires content moderation strategies, such as safety input and
output filters. Here, we provide a theoretical framework for conceptualizing
responsible content moderation of text-to-image generative technologies,
including a demonstration of how to empirically measure the constructs we
enumerate. We define and distinguish the concepts of safety, fairness, and
metric equity, and enumerate example harms that can come in each domain. We
then provide a demonstration of how the defined harms can be quantified. We
conclude with a summary of how the style of harms quantification we demonstrate
enables data-driven content moderation decisions.
|
2306.06139 | Ravindrakumar Purohit Purohit | Ravindrakumar Purohit, Jai Prakash Verma, Rachna Jain, Madhuri Bhavsar | WePaMaDM-Outlier Detection: Weighted Outlier Detection using Pattern
Approaches for Mass Data Mining | null | null | null | null | cs.LG cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Weighted Outlier Detection is a method for identifying unusual or anomalous
data points in a dataset, which can be caused by various factors like human
error, fraud, or equipment malfunctions. Detecting outliers can reveal vital
information about system faults, fraudulent activities, and patterns in the
data, assisting experts in addressing the root causes of these anomalies.
However,creating a model of normal data patterns to identify outliers can be
challenging due to the nature of input data, labeled data availability, and
specific requirements of the problem. This article proposed the
WePaMaDM-Outlier Detection with distinct mass data mining domain, demonstrating
that such techniques are domain-dependent and usually developed for specific
problem formulations. Nevertheless, similar domains can adapt solutions with
modifications. This work also investigates the significance of data modeling in
outlier detection techniques in surveillance, fault detection, and trend
analysis, also referred to as novelty detection, a semisupervised task where
the algorithm learns to recognize abnormality while being taught the normal
class.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 07:00:00 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Purohit",
"Ravindrakumar",
""
],
[
"Verma",
"Jai Prakash",
""
],
[
"Jain",
"Rachna",
""
],
[
"Bhavsar",
"Madhuri",
""
]
] | TITLE: WePaMaDM-Outlier Detection: Weighted Outlier Detection using Pattern
Approaches for Mass Data Mining
ABSTRACT: Weighted Outlier Detection is a method for identifying unusual or anomalous
data points in a dataset, which can be caused by various factors like human
error, fraud, or equipment malfunctions. Detecting outliers can reveal vital
information about system faults, fraudulent activities, and patterns in the
data, assisting experts in addressing the root causes of these anomalies.
However,creating a model of normal data patterns to identify outliers can be
challenging due to the nature of input data, labeled data availability, and
specific requirements of the problem. This article proposed the
WePaMaDM-Outlier Detection with distinct mass data mining domain, demonstrating
that such techniques are domain-dependent and usually developed for specific
problem formulations. Nevertheless, similar domains can adapt solutions with
modifications. This work also investigates the significance of data modeling in
outlier detection techniques in surveillance, fault detection, and trend
analysis, also referred to as novelty detection, a semisupervised task where
the algorithm learns to recognize abnormality while being taught the normal
class.
|
2306.06141 | Ze-Song Xu | Ze-Song Xu, Yun-Nung Chen | Zero-Shot Dialogue Relation Extraction by Relating Explainable Triggers
and Relation Names | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Developing dialogue relation extraction (DRE) systems often requires a large
amount of labeled data, which can be costly and time-consuming to annotate. In
order to improve scalability and support diverse, unseen relation extraction,
this paper proposes a method for leveraging the ability to capture triggers and
relate them to previously unseen relation names. Specifically, we introduce a
model that enables zero-shot dialogue relation extraction by utilizing
trigger-capturing capabilities. Our experiments on a benchmark DialogRE dataset
demonstrate that the proposed model achieves significant improvements for both
seen and unseen relations. Notably, this is the first attempt at zero-shot
dialogue relation extraction using trigger-capturing capabilities, and our
results suggest that this approach is effective for inferring previously unseen
relation types. Overall, our findings highlight the potential for this method
to enhance the scalability and practicality of DRE systems.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 07:10:01 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Xu",
"Ze-Song",
""
],
[
"Chen",
"Yun-Nung",
""
]
] | TITLE: Zero-Shot Dialogue Relation Extraction by Relating Explainable Triggers
and Relation Names
ABSTRACT: Developing dialogue relation extraction (DRE) systems often requires a large
amount of labeled data, which can be costly and time-consuming to annotate. In
order to improve scalability and support diverse, unseen relation extraction,
this paper proposes a method for leveraging the ability to capture triggers and
relate them to previously unseen relation names. Specifically, we introduce a
model that enables zero-shot dialogue relation extraction by utilizing
trigger-capturing capabilities. Our experiments on a benchmark DialogRE dataset
demonstrate that the proposed model achieves significant improvements for both
seen and unseen relations. Notably, this is the first attempt at zero-shot
dialogue relation extraction using trigger-capturing capabilities, and our
results suggest that this approach is effective for inferring previously unseen
relation types. Overall, our findings highlight the potential for this method
to enhance the scalability and practicality of DRE systems.
|
2306.06142 | Nathan Doumeche | Yvenn Amara-Ouali (EDF R&D), Yannig Goude (EDF R&D), Nathan Doum\`eche
(SU, EDF R&D), Pascal Veyret (EDF R&D), Alexis Thomas, Daniel Hebenstreit (TU
Graz), Thomas Wedenig (TU Graz), Arthur Satouf, Aymeric Jan, Yannick Deleuze
(VeRI), Paul Berhaut, S\'ebastien Treguer, Tiphaine Phe-Neau | Forecasting Electric Vehicle Charging Station Occupancy: Smarter
Mobility Data Challenge | null | null | null | null | cs.DB stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The transport sector is a major contributor to greenhouse gas emissions in
Europe. Shifting to electric vehicles (EVs) powered by a low-carbon energy mix
would reduce carbon emissions. However, to support the development of electric
mobility, a better understanding of EV charging behaviours and more accurate
forecasting models are needed. To fill that gap, the Smarter Mobility Data
Challenge has focused on the development of forecasting models to predict EV
charging station occupancy. This challenge involved analysing a dataset of 91
charging stations across four geographical areas over seven months in
2020-2021. The forecasts were evaluated at three levels of aggregation
(individual stations, areas and global) to capture the inherent hierarchical
structure of the data. The results highlight the potential of hierarchical
forecasting approaches to accurately predict EV charging station occupancy,
providing valuable insights for energy providers and EV users alike. This open
dataset addresses many real-world challenges associated with time series, such
as missing values, non-stationarity and spatio-temporal correlations. Access to
the dataset, code and benchmarks are available at
https://gitlab.com/smarter-mobility-data-challenge/tutorials to foster future
research.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 07:22:18 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Amara-Ouali",
"Yvenn",
"",
"EDF R&D"
],
[
"Goude",
"Yannig",
"",
"EDF R&D"
],
[
"Doumèche",
"Nathan",
"",
"SU, EDF R&D"
],
[
"Veyret",
"Pascal",
"",
"EDF R&D"
],
[
"Thomas",
"Alexis",
"",
"TU\n Graz"
],
[
"Hebenstreit",
"Daniel",
"",
"TU\n Graz"
],
[
"Wedenig",
"Thomas",
"",
"TU Graz"
],
[
"Satouf",
"Arthur",
"",
"VeRI"
],
[
"Jan",
"Aymeric",
"",
"VeRI"
],
[
"Deleuze",
"Yannick",
"",
"VeRI"
],
[
"Berhaut",
"Paul",
""
],
[
"Treguer",
"Sébastien",
""
],
[
"Phe-Neau",
"Tiphaine",
""
]
] | TITLE: Forecasting Electric Vehicle Charging Station Occupancy: Smarter
Mobility Data Challenge
ABSTRACT: The transport sector is a major contributor to greenhouse gas emissions in
Europe. Shifting to electric vehicles (EVs) powered by a low-carbon energy mix
would reduce carbon emissions. However, to support the development of electric
mobility, a better understanding of EV charging behaviours and more accurate
forecasting models are needed. To fill that gap, the Smarter Mobility Data
Challenge has focused on the development of forecasting models to predict EV
charging station occupancy. This challenge involved analysing a dataset of 91
charging stations across four geographical areas over seven months in
2020-2021. The forecasts were evaluated at three levels of aggregation
(individual stations, areas and global) to capture the inherent hierarchical
structure of the data. The results highlight the potential of hierarchical
forecasting approaches to accurately predict EV charging station occupancy,
providing valuable insights for energy providers and EV users alike. This open
dataset addresses many real-world challenges associated with time series, such
as missing values, non-stationarity and spatio-temporal correlations. Access to
the dataset, code and benchmarks are available at
https://gitlab.com/smarter-mobility-data-challenge/tutorials to foster future
research.
|
2306.06147 | Labib Chowdhury | Md. Ekramul Islam, Labib Chowdhury, Faisal Ahamed Khan, Shazzad
Hossain, Sourave Hossain, Mohammad Mamun Or Rashid, Nabeel Mohammed and
Mohammad Ruhul Amin | SentiGOLD: A Large Bangla Gold Standard Multi-Domain Sentiment Analysis
Dataset and its Evaluation | Accepted in KDD 2023 Applied Data Science Track; 12 pages, 14 figures | null | 10.1145/3580305.3599904 | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | This study introduces SentiGOLD, a Bangla multi-domain sentiment analysis
dataset. Comprising 70,000 samples, it was created from diverse sources and
annotated by a gender-balanced team of linguists. SentiGOLD adheres to
established linguistic conventions agreed upon by the Government of Bangladesh
and a Bangla linguistics committee. Unlike English and other languages, Bangla
lacks standard sentiment analysis datasets due to the absence of a national
linguistics framework. The dataset incorporates data from online video
comments, social media posts, blogs, news, and other sources while maintaining
domain and class distribution rigorously. It spans 30 domains (e.g., politics,
entertainment, sports) and includes 5 sentiment classes (strongly negative,
weakly negative, neutral, and strongly positive). The annotation scheme,
approved by the national linguistics committee, ensures a robust Inter
Annotator Agreement (IAA) with a Fleiss' kappa score of 0.88. Intra- and
cross-dataset evaluation protocols are applied to establish a standard
classification system. Cross-dataset evaluation on the noisy SentNoB dataset
presents a challenging test scenario. Additionally, zero-shot experiments
demonstrate the generalizability of SentiGOLD. The top model achieves a macro
f1 score of 0.62 (intra-dataset) across 5 classes, setting a benchmark, and
0.61 (cross-dataset from SentNoB) across 3 classes, comparable to the
state-of-the-art. Fine-tuned sentiment analysis model can be accessed at
https://sentiment.bangla.gov.bd.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 12:07:10 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Islam",
"Md. Ekramul",
""
],
[
"Chowdhury",
"Labib",
""
],
[
"Khan",
"Faisal Ahamed",
""
],
[
"Hossain",
"Shazzad",
""
],
[
"Hossain",
"Sourave",
""
],
[
"Rashid",
"Mohammad Mamun Or",
""
],
[
"Mohammed",
"Nabeel",
""
],
[
"Amin",
"Mohammad Ruhul",
""
]
] | TITLE: SentiGOLD: A Large Bangla Gold Standard Multi-Domain Sentiment Analysis
Dataset and its Evaluation
ABSTRACT: This study introduces SentiGOLD, a Bangla multi-domain sentiment analysis
dataset. Comprising 70,000 samples, it was created from diverse sources and
annotated by a gender-balanced team of linguists. SentiGOLD adheres to
established linguistic conventions agreed upon by the Government of Bangladesh
and a Bangla linguistics committee. Unlike English and other languages, Bangla
lacks standard sentiment analysis datasets due to the absence of a national
linguistics framework. The dataset incorporates data from online video
comments, social media posts, blogs, news, and other sources while maintaining
domain and class distribution rigorously. It spans 30 domains (e.g., politics,
entertainment, sports) and includes 5 sentiment classes (strongly negative,
weakly negative, neutral, and strongly positive). The annotation scheme,
approved by the national linguistics committee, ensures a robust Inter
Annotator Agreement (IAA) with a Fleiss' kappa score of 0.88. Intra- and
cross-dataset evaluation protocols are applied to establish a standard
classification system. Cross-dataset evaluation on the noisy SentNoB dataset
presents a challenging test scenario. Additionally, zero-shot experiments
demonstrate the generalizability of SentiGOLD. The top model achieves a macro
f1 score of 0.62 (intra-dataset) across 5 classes, setting a benchmark, and
0.61 (cross-dataset from SentNoB) across 3 classes, comparable to the
state-of-the-art. Fine-tuned sentiment analysis model can be accessed at
https://sentiment.bangla.gov.bd.
|
2306.06191 | Anthony Cintron Roman | Anthony Cintron Roman, Kevin Xu, Arfon Smith, Jehu Torres Vega, Caleb
Robinson, Juan M Lavista Ferres | Open Data on GitHub: Unlocking the Potential of AI | In submission to NeurIPS 2023 Track Datasets and Benchmarks | null | null | null | cs.LG cs.IR | http://creativecommons.org/licenses/by/4.0/ | GitHub is the world's largest platform for collaborative software
development, with over 100 million users. GitHub is also used extensively for
open data collaboration, hosting more than 800 million open data files,
totaling 142 terabytes of data. This study highlights the potential of open
data on GitHub and demonstrates how it can accelerate AI research. We analyze
the existing landscape of open data on GitHub and the patterns of how users
share datasets. Our findings show that GitHub is one of the largest hosts of
open data in the world and has experienced an accelerated growth of open data
assets over the past four years. By examining the open data landscape on
GitHub, we aim to empower users and organizations to leverage existing open
datasets and improve their discoverability -- ultimately contributing to the
ongoing AI revolution to help address complex societal issues. We release the
three datasets that we have collected to support this analysis as open datasets
at https://github.com/github/open-data-on-github.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 18:43:26 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Roman",
"Anthony Cintron",
""
],
[
"Xu",
"Kevin",
""
],
[
"Smith",
"Arfon",
""
],
[
"Vega",
"Jehu Torres",
""
],
[
"Robinson",
"Caleb",
""
],
[
"Ferres",
"Juan M Lavista",
""
]
] | TITLE: Open Data on GitHub: Unlocking the Potential of AI
ABSTRACT: GitHub is the world's largest platform for collaborative software
development, with over 100 million users. GitHub is also used extensively for
open data collaboration, hosting more than 800 million open data files,
totaling 142 terabytes of data. This study highlights the potential of open
data on GitHub and demonstrates how it can accelerate AI research. We analyze
the existing landscape of open data on GitHub and the patterns of how users
share datasets. Our findings show that GitHub is one of the largest hosts of
open data in the world and has experienced an accelerated growth of open data
assets over the past four years. By examining the open data landscape on
GitHub, we aim to empower users and organizations to leverage existing open
datasets and improve their discoverability -- ultimately contributing to the
ongoing AI revolution to help address complex societal issues. We release the
three datasets that we have collected to support this analysis as open datasets
at https://github.com/github/open-data-on-github.
|
2306.06212 | Ian Huang | Ian Huang, Vrishab Krishna, Omoruyi Atekha, Leonidas Guibas | Aladdin: Zero-Shot Hallucination of Stylized 3D Assets from Abstract
Scene Descriptions | null | null | null | null | cs.CV cs.GR | http://creativecommons.org/licenses/by/4.0/ | What constitutes the "vibe" of a particular scene? What should one find in "a
busy, dirty city street", "an idyllic countryside", or "a crime scene in an
abandoned living room"? The translation from abstract scene descriptions to
stylized scene elements cannot be done with any generality by extant systems
trained on rigid and limited indoor datasets. In this paper, we propose to
leverage the knowledge captured by foundation models to accomplish this
translation. We present a system that can serve as a tool to generate stylized
assets for 3D scenes described by a short phrase, without the need to enumerate
the objects to be found within the scene or give instructions on their
appearance. Additionally, it is robust to open-world concepts in a way that
traditional methods trained on limited data are not, affording more creative
freedom to the 3D artist. Our system demonstrates this using a foundation model
"team" composed of a large language model, a vision-language model and several
image diffusion models, which communicate using an interpretable and
user-editable intermediate representation, thus allowing for more versatile and
controllable stylized asset generation for 3D artists. We introduce novel
metrics for this task, and show through human evaluations that in 91% of the
cases, our system outputs are judged more faithful to the semantics of the
input scene description than the baseline, thus highlighting the potential of
this approach to radically accelerate the 3D content creation process for 3D
artists.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 19:24:39 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Huang",
"Ian",
""
],
[
"Krishna",
"Vrishab",
""
],
[
"Atekha",
"Omoruyi",
""
],
[
"Guibas",
"Leonidas",
""
]
] | TITLE: Aladdin: Zero-Shot Hallucination of Stylized 3D Assets from Abstract
Scene Descriptions
ABSTRACT: What constitutes the "vibe" of a particular scene? What should one find in "a
busy, dirty city street", "an idyllic countryside", or "a crime scene in an
abandoned living room"? The translation from abstract scene descriptions to
stylized scene elements cannot be done with any generality by extant systems
trained on rigid and limited indoor datasets. In this paper, we propose to
leverage the knowledge captured by foundation models to accomplish this
translation. We present a system that can serve as a tool to generate stylized
assets for 3D scenes described by a short phrase, without the need to enumerate
the objects to be found within the scene or give instructions on their
appearance. Additionally, it is robust to open-world concepts in a way that
traditional methods trained on limited data are not, affording more creative
freedom to the 3D artist. Our system demonstrates this using a foundation model
"team" composed of a large language model, a vision-language model and several
image diffusion models, which communicate using an interpretable and
user-editable intermediate representation, thus allowing for more versatile and
controllable stylized asset generation for 3D artists. We introduce novel
metrics for this task, and show through human evaluations that in 91% of the
cases, our system outputs are judged more faithful to the semantics of the
input scene description than the baseline, thus highlighting the potential of
this approach to radically accelerate the 3D content creation process for 3D
artists.
|
2306.06217 | Saber Mirzaee Bafti | Saber Mirzaee Bafti, Chee Siang Ang, Gianluca Marcelli, Md. Moinul
Hossain, Sadiya Maxamhud, Anastasios D. Tsaousis | BioGAN: An unpaired GAN-based image to image translation model for
microbiological images | null | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | A diversified dataset is crucial for training a well-generalized supervised
computer vision algorithm. However, in the field of microbiology, generation
and annotation of a diverse dataset including field-taken images are time
consuming, costly, and in some cases impossible. Image to image translation
frameworks allow us to diversify the dataset by transferring images from one
domain to another. However, most existing image translation techniques require
a paired dataset (original image and its corresponding image in the target
domain), which poses a significant challenge in collecting such datasets. In
addition, the application of these image translation frameworks in microbiology
is rarely discussed. In this study, we aim to develop an unpaired GAN-based
(Generative Adversarial Network) image to image translation model for
microbiological images, and study how it can improve generalization ability of
object detection models. In this paper, we present an unpaired and unsupervised
image translation model to translate laboratory-taken microbiological images to
field images, building upon the recent advances in GAN networks and Perceptual
loss function. We propose a novel design for a GAN model, BioGAN, by utilizing
Adversarial and Perceptual loss in order to transform high level features of
laboratory-taken images into field images, while keeping their spatial
features. The contribution of Adversarial and Perceptual loss in the generation
of realistic field images were studied. We used the synthetic field images,
generated by BioGAN, to train an object-detection framework, and compared the
results with those of an object-detection framework trained with laboratory
images; this resulted in up to 68.1% and 75.3% improvement on F1-score and mAP,
respectively. Codes is publicly available at
https://github.com/Kahroba2000/BioGAN.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 19:30:49 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Bafti",
"Saber Mirzaee",
""
],
[
"Ang",
"Chee Siang",
""
],
[
"Marcelli",
"Gianluca",
""
],
[
"Hossain",
"Md. Moinul",
""
],
[
"Maxamhud",
"Sadiya",
""
],
[
"Tsaousis",
"Anastasios D.",
""
]
] | TITLE: BioGAN: An unpaired GAN-based image to image translation model for
microbiological images
ABSTRACT: A diversified dataset is crucial for training a well-generalized supervised
computer vision algorithm. However, in the field of microbiology, generation
and annotation of a diverse dataset including field-taken images are time
consuming, costly, and in some cases impossible. Image to image translation
frameworks allow us to diversify the dataset by transferring images from one
domain to another. However, most existing image translation techniques require
a paired dataset (original image and its corresponding image in the target
domain), which poses a significant challenge in collecting such datasets. In
addition, the application of these image translation frameworks in microbiology
is rarely discussed. In this study, we aim to develop an unpaired GAN-based
(Generative Adversarial Network) image to image translation model for
microbiological images, and study how it can improve generalization ability of
object detection models. In this paper, we present an unpaired and unsupervised
image translation model to translate laboratory-taken microbiological images to
field images, building upon the recent advances in GAN networks and Perceptual
loss function. We propose a novel design for a GAN model, BioGAN, by utilizing
Adversarial and Perceptual loss in order to transform high level features of
laboratory-taken images into field images, while keeping their spatial
features. The contribution of Adversarial and Perceptual loss in the generation
of realistic field images were studied. We used the synthetic field images,
generated by BioGAN, to train an object-detection framework, and compared the
results with those of an object-detection framework trained with laboratory
images; this resulted in up to 68.1% and 75.3% improvement on F1-score and mAP,
respectively. Codes is publicly available at
https://github.com/Kahroba2000/BioGAN.
|
2306.06228 | Robert Joyce | Robert J. Joyce, Tirth Patel, Charles Nicholas, Edward Raff | AVScan2Vec: Feature Learning on Antivirus Scan Data for Production-Scale
Malware Corpora | null | null | null | null | cs.CR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | When investigating a malicious file, searching for related files is a common
task that malware analysts must perform. Given that production malware corpora
may contain over a billion files and consume petabytes of storage, many feature
extraction and similarity search approaches are computationally infeasible. Our
work explores the potential of antivirus (AV) scan data as a scalable source of
features for malware. This is possible because AV scan reports are widely
available through services such as VirusTotal and are ~100x smaller than the
average malware sample. The information within an AV scan report is abundant
with information and can indicate a malicious file's family, behavior, target
operating system, and many other characteristics. We introduce AVScan2Vec, a
language model trained to comprehend the semantics of AV scan data. AVScan2Vec
ingests AV scan data for a malicious file and outputs a meaningful vector
representation. AVScan2Vec vectors are ~3 to 85x smaller than popular
alternatives in use today, enabling faster vector comparisons and lower memory
usage. By incorporating Dynamic Continuous Indexing, we show that
nearest-neighbor queries on AVScan2Vec vectors can scale to even the largest
malware production datasets. We also demonstrate that AVScan2Vec vectors are
superior to other leading malware feature vector representations across nearly
all classification, clustering, and nearest-neighbor lookup algorithms that we
evaluated.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 19:53:40 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Joyce",
"Robert J.",
""
],
[
"Patel",
"Tirth",
""
],
[
"Nicholas",
"Charles",
""
],
[
"Raff",
"Edward",
""
]
] | TITLE: AVScan2Vec: Feature Learning on Antivirus Scan Data for Production-Scale
Malware Corpora
ABSTRACT: When investigating a malicious file, searching for related files is a common
task that malware analysts must perform. Given that production malware corpora
may contain over a billion files and consume petabytes of storage, many feature
extraction and similarity search approaches are computationally infeasible. Our
work explores the potential of antivirus (AV) scan data as a scalable source of
features for malware. This is possible because AV scan reports are widely
available through services such as VirusTotal and are ~100x smaller than the
average malware sample. The information within an AV scan report is abundant
with information and can indicate a malicious file's family, behavior, target
operating system, and many other characteristics. We introduce AVScan2Vec, a
language model trained to comprehend the semantics of AV scan data. AVScan2Vec
ingests AV scan data for a malicious file and outputs a meaningful vector
representation. AVScan2Vec vectors are ~3 to 85x smaller than popular
alternatives in use today, enabling faster vector comparisons and lower memory
usage. By incorporating Dynamic Continuous Indexing, we show that
nearest-neighbor queries on AVScan2Vec vectors can scale to even the largest
malware production datasets. We also demonstrate that AVScan2Vec vectors are
superior to other leading malware feature vector representations across nearly
all classification, clustering, and nearest-neighbor lookup algorithms that we
evaluated.
|
2306.06252 | Jerry Yao-Chieh Hu | Alex Reneau, Jerry Yao-Chieh Hu, Chenwei Xu, Weijian Li, Ammar Gilani,
Han Liu | Feature Programming for Multivariate Time Series Prediction | 21 pages, accepted to ICML2023. Code is available at
https://github.com/SirAlex900/FeatureProgramming | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We introduce the concept of programmable feature engineering for time series
modeling and propose a feature programming framework. This framework generates
large amounts of predictive features for noisy multivariate time series while
allowing users to incorporate their inductive bias with minimal effort. The key
motivation of our framework is to view any multivariate time series as a
cumulative sum of fine-grained trajectory increments, with each increment
governed by a novel spin-gas dynamical Ising model. This fine-grained
perspective motivates the development of a parsimonious set of operators that
summarize multivariate time series in an abstract fashion, serving as the
foundation for large-scale automated feature engineering. Numerically, we
validate the efficacy of our method on several synthetic and real-world noisy
time series datasets.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 20:46:55 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Reneau",
"Alex",
""
],
[
"Hu",
"Jerry Yao-Chieh",
""
],
[
"Xu",
"Chenwei",
""
],
[
"Li",
"Weijian",
""
],
[
"Gilani",
"Ammar",
""
],
[
"Liu",
"Han",
""
]
] | TITLE: Feature Programming for Multivariate Time Series Prediction
ABSTRACT: We introduce the concept of programmable feature engineering for time series
modeling and propose a feature programming framework. This framework generates
large amounts of predictive features for noisy multivariate time series while
allowing users to incorporate their inductive bias with minimal effort. The key
motivation of our framework is to view any multivariate time series as a
cumulative sum of fine-grained trajectory increments, with each increment
governed by a novel spin-gas dynamical Ising model. This fine-grained
perspective motivates the development of a parsimonious set of operators that
summarize multivariate time series in an abstract fashion, serving as the
foundation for large-scale automated feature engineering. Numerically, we
validate the efficacy of our method on several synthetic and real-world noisy
time series datasets.
|
2306.06255 | Zahrasadat Jamadi | Zahra Jamadi and Amir G. Aghdam | Early Malware Detection and Next-Action Prediction | null | null | null | null | cs.CR cs.SY eess.SY | http://creativecommons.org/licenses/by-sa/4.0/ | In this paper, we propose a framework for early-stage malware detection and
mitigation by leveraging natural language processing (NLP) techniques and
machine learning algorithms. Our primary contribution is presenting an approach
for predicting the upcoming actions of malware by treating application
programming interface (API) call sequences as natural language inputs and
employing text classification methods, specifically a Bi-LSTM neural network,
to predict the next API call. This enables proactive threat identification and
mitigation, demonstrating the effectiveness of applying NLP principles to API
call sequences. The Bi-LSTM model is evaluated using two datasets. %The model
achieved an accuracy of 93.6\% and 88.8\% for the %first and second dataset
respectively. Additionally, by modeling consecutive API calls as 2-gram and
3-gram strings, we extract new features to be further processed using a
Bagging-XGBoost algorithm, effectively predicting malware presence at its early
stages. The accuracy of the proposed framework is evaluated by simulations.
| [
{
"version": "v1",
"created": "Fri, 9 Jun 2023 20:57:27 GMT"
}
] | 2023-06-13T00:00:00 | [
[
"Jamadi",
"Zahra",
""
],
[
"Aghdam",
"Amir G.",
""
]
] | TITLE: Early Malware Detection and Next-Action Prediction
ABSTRACT: In this paper, we propose a framework for early-stage malware detection and
mitigation by leveraging natural language processing (NLP) techniques and
machine learning algorithms. Our primary contribution is presenting an approach
for predicting the upcoming actions of malware by treating application
programming interface (API) call sequences as natural language inputs and
employing text classification methods, specifically a Bi-LSTM neural network,
to predict the next API call. This enables proactive threat identification and
mitigation, demonstrating the effectiveness of applying NLP principles to API
call sequences. The Bi-LSTM model is evaluated using two datasets. %The model
achieved an accuracy of 93.6\% and 88.8\% for the %first and second dataset
respectively. Additionally, by modeling consecutive API calls as 2-gram and
3-gram strings, we extract new features to be further processed using a
Bagging-XGBoost algorithm, effectively predicting malware presence at its early
stages. The accuracy of the proposed framework is evaluated by simulations.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.