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
| label
stringclasses 2
values | prob
float64 0.5
0.73
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2207.00012 | Kuan Li | Kuan Li, Yang Liu, Xiang Ao, Jianfeng Chi, Jinghua Feng, Hao Yang,
Qing He | Reliable Representations Make A Stronger Defender: Unsupervised
Structure Refinement for Robust GNN | Accepted in KDD2022 | null | 10.1145/3534678.3539484 | null | cs.LG cs.CR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Benefiting from the message passing mechanism, Graph Neural Networks (GNNs)
have been successful on flourish tasks over graph data. However, recent studies
have shown that attackers can catastrophically degrade the performance of GNNs
by maliciously modifying the graph structure. A straightforward solution to
remedy this issue is to model the edge weights by learning a metric function
between pairwise representations of two end nodes, which attempts to assign low
weights to adversarial edges. The existing methods use either raw features or
representations learned by supervised GNNs to model the edge weights. However,
both strategies are faced with some immediate problems: raw features cannot
represent various properties of nodes (e.g., structure information), and
representations learned by supervised GNN may suffer from the poor performance
of the classifier on the poisoned graph. We need representations that carry
both feature information and as mush correct structure information as possible
and are insensitive to structural perturbations. To this end, we propose an
unsupervised pipeline, named STABLE, to optimize the graph structure. Finally,
we input the well-refined graph into a downstream classifier. For this part, we
design an advanced GCN that significantly enhances the robustness of vanilla
GCN without increasing the time complexity. Extensive experiments on four
real-world graph benchmarks demonstrate that STABLE outperforms the
state-of-the-art methods and successfully defends against various attacks.
| [
{
"version": "v1",
"created": "Thu, 30 Jun 2022 10:02:32 GMT"
},
{
"version": "v2",
"created": "Fri, 2 Sep 2022 03:26:56 GMT"
},
{
"version": "v3",
"created": "Tue, 27 Sep 2022 09:38:07 GMT"
},
{
"version": "v4",
"created": "Fri, 21 Apr 2023 09:01:42 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Li",
"Kuan",
""
],
[
"Liu",
"Yang",
""
],
[
"Ao",
"Xiang",
""
],
[
"Chi",
"Jianfeng",
""
],
[
"Feng",
"Jinghua",
""
],
[
"Yang",
"Hao",
""
],
[
"He",
"Qing",
""
]
] | TITLE: Reliable Representations Make A Stronger Defender: Unsupervised
Structure Refinement for Robust GNN
ABSTRACT: Benefiting from the message passing mechanism, Graph Neural Networks (GNNs)
have been successful on flourish tasks over graph data. However, recent studies
have shown that attackers can catastrophically degrade the performance of GNNs
by maliciously modifying the graph structure. A straightforward solution to
remedy this issue is to model the edge weights by learning a metric function
between pairwise representations of two end nodes, which attempts to assign low
weights to adversarial edges. The existing methods use either raw features or
representations learned by supervised GNNs to model the edge weights. However,
both strategies are faced with some immediate problems: raw features cannot
represent various properties of nodes (e.g., structure information), and
representations learned by supervised GNN may suffer from the poor performance
of the classifier on the poisoned graph. We need representations that carry
both feature information and as mush correct structure information as possible
and are insensitive to structural perturbations. To this end, we propose an
unsupervised pipeline, named STABLE, to optimize the graph structure. Finally,
we input the well-refined graph into a downstream classifier. For this part, we
design an advanced GCN that significantly enhances the robustness of vanilla
GCN without increasing the time complexity. Extensive experiments on four
real-world graph benchmarks demonstrate that STABLE outperforms the
state-of-the-art methods and successfully defends against various attacks.
| no_new_dataset | 0.707998 |
2207.05013 | Alexey Korovin | Alexey N. Korovin, Innokentiy S. Humonen, Artem I. Samtsevich, Roman
A. Eremin, Artem I. Vasilyev, Vladimir D. Lazarev, Semen A. Budennyy | Boosting Heterogeneous Catalyst Discovery by Structurally Constrained
Deep Learning Models | null | Materials Today Chemistry 2023, 30, 101541 | 10.1016/j.mtchem.2023.101541 | null | cond-mat.mtrl-sci cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The discovery of new catalysts is one of the significant topics of
computational chemistry as it has the potential to accelerate the adoption of
renewable energy sources. Recently developed deep learning approaches such as
graph neural networks (GNNs) open new opportunity to significantly extend scope
for modelling novel high-performance catalysts. Nevertheless, the graph
representation of particular crystal structure is not a straightforward task
due to the ambiguous connectivity schemes and numerous embeddings of nodes and
edges. Here we present embedding improvement for GNN that has been modified by
Voronoi tesselation and is able to predict the energy of catalytic systems
within Open Catalyst Project dataset. Enrichment of the graph was calculated
via Voronoi tessellation and the corresponding contact solid angles and types
(direct or indirect) were considered as features of edges and Voronoi volumes
were used as node characteristics. The auxiliary approach was enriching node
representation by intrinsic atomic properties (electronegativity, period and
group position). Proposed modifications allowed us to improve the mean absolute
error of the original model and the final error equals to 651 meV per atom on
the Open Catalyst Project dataset and 6 meV per atom on the intermetallics
dataset. Also, by consideration of additional dataset, we show that a sensible
choice of data can decrease the error to values above physically-based 20 meV
per atom threshold.
| [
{
"version": "v1",
"created": "Mon, 11 Jul 2022 17:01:28 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Sep 2022 13:49:45 GMT"
},
{
"version": "v3",
"created": "Sun, 2 Oct 2022 15:30:48 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Korovin",
"Alexey N.",
""
],
[
"Humonen",
"Innokentiy S.",
""
],
[
"Samtsevich",
"Artem I.",
""
],
[
"Eremin",
"Roman A.",
""
],
[
"Vasilyev",
"Artem I.",
""
],
[
"Lazarev",
"Vladimir D.",
""
],
[
"Budennyy",
"Semen A.",
""
]
] | TITLE: Boosting Heterogeneous Catalyst Discovery by Structurally Constrained
Deep Learning Models
ABSTRACT: The discovery of new catalysts is one of the significant topics of
computational chemistry as it has the potential to accelerate the adoption of
renewable energy sources. Recently developed deep learning approaches such as
graph neural networks (GNNs) open new opportunity to significantly extend scope
for modelling novel high-performance catalysts. Nevertheless, the graph
representation of particular crystal structure is not a straightforward task
due to the ambiguous connectivity schemes and numerous embeddings of nodes and
edges. Here we present embedding improvement for GNN that has been modified by
Voronoi tesselation and is able to predict the energy of catalytic systems
within Open Catalyst Project dataset. Enrichment of the graph was calculated
via Voronoi tessellation and the corresponding contact solid angles and types
(direct or indirect) were considered as features of edges and Voronoi volumes
were used as node characteristics. The auxiliary approach was enriching node
representation by intrinsic atomic properties (electronegativity, period and
group position). Proposed modifications allowed us to improve the mean absolute
error of the original model and the final error equals to 651 meV per atom on
the Open Catalyst Project dataset and 6 meV per atom on the intermetallics
dataset. Also, by consideration of additional dataset, we show that a sensible
choice of data can decrease the error to values above physically-based 20 meV
per atom threshold.
| no_new_dataset | 0.710025 |
2209.08473 | Shitong Shao | Huanran Chen, Shitong Shao, Ziyi Wang, Zirui Shang, Jin Chen, Xiaofeng
Ji, Xinxiao Wu | Bootstrap Generalization Ability from Loss Landscape Perspective | 18 pages, 4 figures, Accepted by ECCV Workshop2022 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Domain generalization aims to learn a model that can generalize well on the
unseen test dataset, i.e., out-of-distribution data, which has different
distribution from the training dataset. To address domain generalization in
computer vision, we introduce the loss landscape theory into this field.
Specifically, we bootstrap the generalization ability of the deep learning
model from the loss landscape perspective in four aspects, including backbone,
regularization, training paradigm, and learning rate. We verify the proposed
theory on the NICO++, PACS, and VLCS datasets by doing extensive ablation
studies as well as visualizations. In addition, we apply this theory in the
ECCV 2022 NICO Challenge1 and achieve the 3rd place without using any domain
invariant methods.
| [
{
"version": "v1",
"created": "Sun, 18 Sep 2022 04:40:32 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 12:45:27 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Chen",
"Huanran",
""
],
[
"Shao",
"Shitong",
""
],
[
"Wang",
"Ziyi",
""
],
[
"Shang",
"Zirui",
""
],
[
"Chen",
"Jin",
""
],
[
"Ji",
"Xiaofeng",
""
],
[
"Wu",
"Xinxiao",
""
]
] | TITLE: Bootstrap Generalization Ability from Loss Landscape Perspective
ABSTRACT: Domain generalization aims to learn a model that can generalize well on the
unseen test dataset, i.e., out-of-distribution data, which has different
distribution from the training dataset. To address domain generalization in
computer vision, we introduce the loss landscape theory into this field.
Specifically, we bootstrap the generalization ability of the deep learning
model from the loss landscape perspective in four aspects, including backbone,
regularization, training paradigm, and learning rate. We verify the proposed
theory on the NICO++, PACS, and VLCS datasets by doing extensive ablation
studies as well as visualizations. In addition, we apply this theory in the
ECCV 2022 NICO Challenge1 and achieve the 3rd place without using any domain
invariant methods.
| no_new_dataset | 0.712589 |
2209.12396 | Pengxin Zeng | Pengxin Zeng, Yunfan Li, Peng Hu, Dezhong Peng, Jiancheng Lv, Xi Peng | Deep Fair Clustering via Maximizing and Minimizing Mutual Information:
Theory, Algorithm and Metric | null | null | null | null | cs.LG cs.CY | http://creativecommons.org/licenses/by-sa/4.0/ | Fair clustering aims to divide data into distinct clusters while preventing
sensitive attributes (\textit{e.g.}, gender, race, RNA sequencing technique)
from dominating the clustering. Although a number of works have been conducted
and achieved huge success recently, most of them are heuristical, and there
lacks a unified theory for algorithm design. In this work, we fill this blank
by developing a mutual information theory for deep fair clustering and
accordingly designing a novel algorithm, dubbed FCMI. In brief, through
maximizing and minimizing mutual information, FCMI is designed to achieve four
characteristics highly expected by deep fair clustering, \textit{i.e.},
compact, balanced, and fair clusters, as well as informative features. Besides
the contributions to theory and algorithm, another contribution of this work is
proposing a novel fair clustering metric built upon information theory as well.
Unlike existing evaluation metrics, our metric measures the clustering quality
and fairness as a whole instead of separate manner. To verify the effectiveness
of the proposed FCMI, we conduct experiments on six benchmarks including a
single-cell RNA-seq atlas compared with 11 state-of-the-art methods in terms of
five metrics. The code could be accessed from \url{ https://pengxi.me}.
| [
{
"version": "v1",
"created": "Mon, 26 Sep 2022 03:38:48 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 04:07:41 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Zeng",
"Pengxin",
""
],
[
"Li",
"Yunfan",
""
],
[
"Hu",
"Peng",
""
],
[
"Peng",
"Dezhong",
""
],
[
"Lv",
"Jiancheng",
""
],
[
"Peng",
"Xi",
""
]
] | TITLE: Deep Fair Clustering via Maximizing and Minimizing Mutual Information:
Theory, Algorithm and Metric
ABSTRACT: Fair clustering aims to divide data into distinct clusters while preventing
sensitive attributes (\textit{e.g.}, gender, race, RNA sequencing technique)
from dominating the clustering. Although a number of works have been conducted
and achieved huge success recently, most of them are heuristical, and there
lacks a unified theory for algorithm design. In this work, we fill this blank
by developing a mutual information theory for deep fair clustering and
accordingly designing a novel algorithm, dubbed FCMI. In brief, through
maximizing and minimizing mutual information, FCMI is designed to achieve four
characteristics highly expected by deep fair clustering, \textit{i.e.},
compact, balanced, and fair clusters, as well as informative features. Besides
the contributions to theory and algorithm, another contribution of this work is
proposing a novel fair clustering metric built upon information theory as well.
Unlike existing evaluation metrics, our metric measures the clustering quality
and fairness as a whole instead of separate manner. To verify the effectiveness
of the proposed FCMI, we conduct experiments on six benchmarks including a
single-cell RNA-seq atlas compared with 11 state-of-the-art methods in terms of
five metrics. The code could be accessed from \url{ https://pengxi.me}.
| no_new_dataset | 0.710459 |
2210.01447 | Mansi Sharma | Sally Khaidem and Mansi Sharma | A Novel Light Field Coding Scheme Based on Deep Belief Network &
Weighted Binary Images for Additive Layered Displays | The paper is under consideration at Pattern Recognition Letters | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Light-field displays create an immersive experience by providing binocular
depth sensation and motion parallax. Stacking light attenuating layers is one
approach to implement a light field display with a broader depth of field, wide
viewing angles and high resolution. Due to the transparent holographic optical
element (HOE) layers, additive layered displays can be integrated into
augmented reality (AR) wearables to overlay virtual objects onto the real
world, creating a seamless mixed reality (XR) experience. This paper proposes a
novel framework for light field representation and coding that utilizes Deep
Belief Network (DBN) and weighted binary images suitable for additive layered
displays. The weighted binary representation of layers makes the framework more
flexible for adaptive bitrate encoding. The framework effectively captures
intrinsic redundancies in the light field data, and thus provides a scalable
solution for light field coding suitable for XR display applications. The
latent code is encoded by H.265 codec generating a rate-scalable bit-stream. We
achieve adaptive bitrate decoding by varying the number of weighted binary
images and the H.265 quantization parameter, while maintaining an optimal
reconstruction quality. The framework is tested on real and synthetic benchmark
datasets, and the results validate the rate-scalable property of the proposed
scheme.
| [
{
"version": "v1",
"created": "Tue, 4 Oct 2022 08:18:06 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 14:59:26 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Khaidem",
"Sally",
""
],
[
"Sharma",
"Mansi",
""
]
] | TITLE: A Novel Light Field Coding Scheme Based on Deep Belief Network &
Weighted Binary Images for Additive Layered Displays
ABSTRACT: Light-field displays create an immersive experience by providing binocular
depth sensation and motion parallax. Stacking light attenuating layers is one
approach to implement a light field display with a broader depth of field, wide
viewing angles and high resolution. Due to the transparent holographic optical
element (HOE) layers, additive layered displays can be integrated into
augmented reality (AR) wearables to overlay virtual objects onto the real
world, creating a seamless mixed reality (XR) experience. This paper proposes a
novel framework for light field representation and coding that utilizes Deep
Belief Network (DBN) and weighted binary images suitable for additive layered
displays. The weighted binary representation of layers makes the framework more
flexible for adaptive bitrate encoding. The framework effectively captures
intrinsic redundancies in the light field data, and thus provides a scalable
solution for light field coding suitable for XR display applications. The
latent code is encoded by H.265 codec generating a rate-scalable bit-stream. We
achieve adaptive bitrate decoding by varying the number of weighted binary
images and the H.265 quantization parameter, while maintaining an optimal
reconstruction quality. The framework is tested on real and synthetic benchmark
datasets, and the results validate the rate-scalable property of the proposed
scheme.
| no_new_dataset | 0.714392 |
2210.04468 | Ru Peng | Ru Peng, Yawen Zeng, Junbo Zhao | Distill the Image to Nowhere: Inversion Knowledge Distillation for
Multimodal Machine Translation | EMNLP2022 Oral Long paper | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Past works on multimodal machine translation (MMT) elevate bilingual setup by
incorporating additional aligned vision information. However, an image-must
requirement of the multimodal dataset largely hinders MMT's development --
namely that it demands an aligned form of [image, source text, target text].
This limitation is generally troublesome during the inference phase especially
when the aligned image is not provided as in the normal NMT setup. Thus, in
this work, we introduce IKD-MMT, a novel MMT framework to support the
image-free inference phase via an inversion knowledge distillation scheme. In
particular, a multimodal feature generator is executed with a knowledge
distillation module, which directly generates the multimodal feature from
(only) source texts as the input. While there have been a few prior works
entertaining the possibility to support image-free inference for machine
translation, their performances have yet to rival the image-must translation.
In our experiments, we identify our method as the first image-free approach to
comprehensively rival or even surpass (almost) all image-must frameworks, and
achieved the state-of-the-art result on the often-used Multi30k benchmark. Our
code and data are available at: https://github.com/pengr/IKD-mmt/tree/master..
| [
{
"version": "v1",
"created": "Mon, 10 Oct 2022 07:36:59 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 09:40:09 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Peng",
"Ru",
""
],
[
"Zeng",
"Yawen",
""
],
[
"Zhao",
"Junbo",
""
]
] | TITLE: Distill the Image to Nowhere: Inversion Knowledge Distillation for
Multimodal Machine Translation
ABSTRACT: Past works on multimodal machine translation (MMT) elevate bilingual setup by
incorporating additional aligned vision information. However, an image-must
requirement of the multimodal dataset largely hinders MMT's development --
namely that it demands an aligned form of [image, source text, target text].
This limitation is generally troublesome during the inference phase especially
when the aligned image is not provided as in the normal NMT setup. Thus, in
this work, we introduce IKD-MMT, a novel MMT framework to support the
image-free inference phase via an inversion knowledge distillation scheme. In
particular, a multimodal feature generator is executed with a knowledge
distillation module, which directly generates the multimodal feature from
(only) source texts as the input. While there have been a few prior works
entertaining the possibility to support image-free inference for machine
translation, their performances have yet to rival the image-must translation.
In our experiments, we identify our method as the first image-free approach to
comprehensively rival or even surpass (almost) all image-must frameworks, and
achieved the state-of-the-art result on the often-used Multi30k benchmark. Our
code and data are available at: https://github.com/pengr/IKD-mmt/tree/master..
| no_new_dataset | 0.706551 |
2211.05530 | Jan Butora | Jan Butora and Patrick Bas | Side-Informed Steganography for JPEG Images by Modeling Decompressed
Images | 13 pages, 7 figures, 1 table, submitted to IEEE Transactions on
Information Forensics & Security | null | 10.1109/TIFS.2023.3268884 | null | cs.MM cs.CR eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Side-informed steganography has always been among the most secure approaches
in the field. However, a majority of existing methods for JPEG images use the
side information, here the rounding error, in a heuristic way. For the first
time, we show that the usefulness of the rounding error comes from its
covariance with the embedding changes. Unfortunately, this covariance between
continuous and discrete variables is not analytically available. An estimate of
the covariance is proposed, which allows to model steganography as a change in
the variance of DCT coefficients. Since steganalysis today is best performed in
the spatial domain, we derive a likelihood ratio test to preserve a model of a
decompressed JPEG image. The proposed method then bounds the power of this test
by minimizing the Kullback-Leibler divergence between the cover and stego
distributions. We experimentally demonstrate in two popular datasets that it
achieves state-of-the-art performance against deep learning detectors.
Moreover, by considering a different pixel variance estimator for images
compressed with Quality Factor 100, even greater improvements are obtained.
| [
{
"version": "v1",
"created": "Wed, 9 Nov 2022 09:34:11 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Butora",
"Jan",
""
],
[
"Bas",
"Patrick",
""
]
] | TITLE: Side-Informed Steganography for JPEG Images by Modeling Decompressed
Images
ABSTRACT: Side-informed steganography has always been among the most secure approaches
in the field. However, a majority of existing methods for JPEG images use the
side information, here the rounding error, in a heuristic way. For the first
time, we show that the usefulness of the rounding error comes from its
covariance with the embedding changes. Unfortunately, this covariance between
continuous and discrete variables is not analytically available. An estimate of
the covariance is proposed, which allows to model steganography as a change in
the variance of DCT coefficients. Since steganalysis today is best performed in
the spatial domain, we derive a likelihood ratio test to preserve a model of a
decompressed JPEG image. The proposed method then bounds the power of this test
by minimizing the Kullback-Leibler divergence between the cover and stego
distributions. We experimentally demonstrate in two popular datasets that it
achieves state-of-the-art performance against deep learning detectors.
Moreover, by considering a different pixel variance estimator for images
compressed with Quality Factor 100, even greater improvements are obtained.
| no_new_dataset | 0.706646 |
2211.07347 | Claudia Bongiovanni | Claudia Bongiovanni and Nikolas Geroliminis and Mor Kaspi | A ride time-oriented scheduling algorithm for dial-a-ride problems | 12 pages, 1 figure | null | null | null | math.OC cs.DS | http://creativecommons.org/licenses/by/4.0/ | This paper offers a new algorithm to efficiently optimize scheduling
decisions for dial-a-ride problems (DARPs), including problem variants
considering electric and autonomous vehicles (e-ADARPs). The scheduling
heuristic, based on linear programming theory, aims at finding minimal user
ride time schedules in polynomial time. The algorithm can either return optimal
feasible routes or it can return incorrect infeasibility declarations, on which
feasibility can be recovered through a specifically-designed heuristic. The
algorithm is furthermore supplemented by a battery management algorithm that
can be used to determine charging decisions for electric and autonomous vehicle
fleets. Timing solutions from the proposed scheduling algorithm are obtained on
millions of routes extracted from DARP and e-ADARP benchmark instances. They
are compared to those obtained from a linear program, as well as to popular
scheduling procedures from the DARP literature. Results show that the proposed
procedure outperforms state-of-the-art scheduling algorithms, both in terms of
compute-efficiency and solution quality.
| [
{
"version": "v1",
"created": "Mon, 14 Nov 2022 13:40:46 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 23:00:56 GMT"
},
{
"version": "v3",
"created": "Fri, 21 Apr 2023 17:22:11 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Bongiovanni",
"Claudia",
""
],
[
"Geroliminis",
"Nikolas",
""
],
[
"Kaspi",
"Mor",
""
]
] | TITLE: A ride time-oriented scheduling algorithm for dial-a-ride problems
ABSTRACT: This paper offers a new algorithm to efficiently optimize scheduling
decisions for dial-a-ride problems (DARPs), including problem variants
considering electric and autonomous vehicles (e-ADARPs). The scheduling
heuristic, based on linear programming theory, aims at finding minimal user
ride time schedules in polynomial time. The algorithm can either return optimal
feasible routes or it can return incorrect infeasibility declarations, on which
feasibility can be recovered through a specifically-designed heuristic. The
algorithm is furthermore supplemented by a battery management algorithm that
can be used to determine charging decisions for electric and autonomous vehicle
fleets. Timing solutions from the proposed scheduling algorithm are obtained on
millions of routes extracted from DARP and e-ADARP benchmark instances. They
are compared to those obtained from a linear program, as well as to popular
scheduling procedures from the DARP literature. Results show that the proposed
procedure outperforms state-of-the-art scheduling algorithms, both in terms of
compute-efficiency and solution quality.
| no_new_dataset | 0.707411 |
2211.07740 | Mark Graham | Mark S. Graham, Walter H.L. Pinaya, Petru-Daniel Tudosiu, Parashkev
Nachev, Sebastien Ourselin, M. Jorge Cardoso | Denoising diffusion models for out-of-distribution detection | null | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | Out-of-distribution detection is crucial to the safe deployment of machine
learning systems. Currently, unsupervised out-of-distribution detection is
dominated by generative-based approaches that make use of estimates of the
likelihood or other measurements from a generative model. Reconstruction-based
methods offer an alternative approach, in which a measure of reconstruction
error is used to determine if a sample is out-of-distribution. However,
reconstruction-based approaches are less favoured, as they require careful
tuning of the model's information bottleneck - such as the size of the latent
dimension - to produce good results. In this work, we exploit the view of
denoising diffusion probabilistic models (DDPM) as denoising autoencoders where
the bottleneck is controlled externally, by means of the amount of noise
applied. We propose to use DDPMs to reconstruct an input that has been noised
to a range of noise levels, and use the resulting multi-dimensional
reconstruction error to classify out-of-distribution inputs. We validate our
approach both on standard computer-vision datasets and on higher dimension
medical datasets. Our approach outperforms not only reconstruction-based
methods, but also state-of-the-art generative-based approaches. Code is
available at https://github.com/marksgraham/ddpm-ood.
| [
{
"version": "v1",
"created": "Mon, 14 Nov 2022 20:35:11 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Nov 2022 22:00:22 GMT"
},
{
"version": "v3",
"created": "Mon, 6 Mar 2023 01:35:21 GMT"
},
{
"version": "v4",
"created": "Thu, 20 Apr 2023 20:09:54 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Graham",
"Mark S.",
""
],
[
"Pinaya",
"Walter H. L.",
""
],
[
"Tudosiu",
"Petru-Daniel",
""
],
[
"Nachev",
"Parashkev",
""
],
[
"Ourselin",
"Sebastien",
""
],
[
"Cardoso",
"M. Jorge",
""
]
] | TITLE: Denoising diffusion models for out-of-distribution detection
ABSTRACT: Out-of-distribution detection is crucial to the safe deployment of machine
learning systems. Currently, unsupervised out-of-distribution detection is
dominated by generative-based approaches that make use of estimates of the
likelihood or other measurements from a generative model. Reconstruction-based
methods offer an alternative approach, in which a measure of reconstruction
error is used to determine if a sample is out-of-distribution. However,
reconstruction-based approaches are less favoured, as they require careful
tuning of the model's information bottleneck - such as the size of the latent
dimension - to produce good results. In this work, we exploit the view of
denoising diffusion probabilistic models (DDPM) as denoising autoencoders where
the bottleneck is controlled externally, by means of the amount of noise
applied. We propose to use DDPMs to reconstruct an input that has been noised
to a range of noise levels, and use the resulting multi-dimensional
reconstruction error to classify out-of-distribution inputs. We validate our
approach both on standard computer-vision datasets and on higher dimension
medical datasets. Our approach outperforms not only reconstruction-based
methods, but also state-of-the-art generative-based approaches. Code is
available at https://github.com/marksgraham/ddpm-ood.
| no_new_dataset | 0.709252 |
2211.10295 | Raghav Kansal | Raghav Kansal and Anni Li and Javier Duarte and Nadezda Chernyavskaya
and Maurizio Pierini and Breno Orzari and Thiago Tomei | Evaluating generative models in high energy physics | 11 pages, 5 figures, 3 tables, and a 5 page appendix | Phys. Rev. D 107, 076017 (2023) | 10.1103/PhysRevD.107.076017 | FERMILAB-PUB-22-872-CMS-PPD | hep-ex cs.LG stat.AP | http://creativecommons.org/licenses/by/4.0/ | There has been a recent explosion in research into machine-learning-based
generative modeling to tackle computational challenges for simulations in high
energy physics (HEP). In order to use such alternative simulators in practice,
we need well-defined metrics to compare different generative models and
evaluate their discrepancy from the true distributions. We present the first
systematic review and investigation into evaluation metrics and their
sensitivity to failure modes of generative models, using the framework of
two-sample goodness-of-fit testing, and their relevance and viability for HEP.
Inspired by previous work in both physics and computer vision, we propose two
new metrics, the Fr\'echet and kernel physics distances (FPD and KPD,
respectively), and perform a variety of experiments measuring their performance
on simple Gaussian-distributed, and simulated high energy jet datasets. We find
FPD, in particular, to be the most sensitive metric to all alternative jet
distributions tested and recommend its adoption, along with the KPD and
Wasserstein distances between individual feature distributions, for evaluating
generative models in HEP. We finally demonstrate the efficacy of these proposed
metrics in evaluating and comparing a novel attention-based generative
adversarial particle transformer to the state-of-the-art message-passing
generative adversarial network jet simulation model. The code for our proposed
metrics is provided in the open source JetNet Python library.
| [
{
"version": "v1",
"created": "Fri, 18 Nov 2022 15:36:28 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 16:14:36 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Kansal",
"Raghav",
""
],
[
"Li",
"Anni",
""
],
[
"Duarte",
"Javier",
""
],
[
"Chernyavskaya",
"Nadezda",
""
],
[
"Pierini",
"Maurizio",
""
],
[
"Orzari",
"Breno",
""
],
[
"Tomei",
"Thiago",
""
]
] | TITLE: Evaluating generative models in high energy physics
ABSTRACT: There has been a recent explosion in research into machine-learning-based
generative modeling to tackle computational challenges for simulations in high
energy physics (HEP). In order to use such alternative simulators in practice,
we need well-defined metrics to compare different generative models and
evaluate their discrepancy from the true distributions. We present the first
systematic review and investigation into evaluation metrics and their
sensitivity to failure modes of generative models, using the framework of
two-sample goodness-of-fit testing, and their relevance and viability for HEP.
Inspired by previous work in both physics and computer vision, we propose two
new metrics, the Fr\'echet and kernel physics distances (FPD and KPD,
respectively), and perform a variety of experiments measuring their performance
on simple Gaussian-distributed, and simulated high energy jet datasets. We find
FPD, in particular, to be the most sensitive metric to all alternative jet
distributions tested and recommend its adoption, along with the KPD and
Wasserstein distances between individual feature distributions, for evaluating
generative models in HEP. We finally demonstrate the efficacy of these proposed
metrics in evaluating and comparing a novel attention-based generative
adversarial particle transformer to the state-of-the-art message-passing
generative adversarial network jet simulation model. The code for our proposed
metrics is provided in the open source JetNet Python library.
| no_new_dataset | 0.709831 |
2212.07016 | Chengzhi Mao | Chengzhi Mao, Scott Geng, Junfeng Yang, Xin Wang, Carl Vondrick | Understanding Zero-Shot Adversarial Robustness for Large-Scale Models | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Pretrained large-scale vision-language models like CLIP have exhibited strong
generalization over unseen tasks. Yet imperceptible adversarial perturbations
can significantly reduce CLIP's performance on new tasks. In this work, we
identify and explore the problem of \emph{adapting large-scale models for
zero-shot adversarial robustness}. We first identify two key factors during
model adaption -- training losses and adaptation methods -- that affect the
model's zero-shot adversarial robustness. We then propose a text-guided
contrastive adversarial training loss, which aligns the text embeddings and the
adversarial visual features with contrastive learning on a small set of
training data. We apply this training loss to two adaption methods, model
finetuning and visual prompt tuning. We find that visual prompt tuning is more
effective in the absence of texts, while finetuning wins in the existence of
text guidance. Overall, our approach significantly improves the zero-shot
adversarial robustness over CLIP, seeing an average improvement of over 31
points over ImageNet and 15 zero-shot datasets. We hope this work can shed
light on understanding the zero-shot adversarial robustness of large-scale
models.
| [
{
"version": "v1",
"created": "Wed, 14 Dec 2022 04:08:56 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 17:08:27 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Mao",
"Chengzhi",
""
],
[
"Geng",
"Scott",
""
],
[
"Yang",
"Junfeng",
""
],
[
"Wang",
"Xin",
""
],
[
"Vondrick",
"Carl",
""
]
] | TITLE: Understanding Zero-Shot Adversarial Robustness for Large-Scale Models
ABSTRACT: Pretrained large-scale vision-language models like CLIP have exhibited strong
generalization over unseen tasks. Yet imperceptible adversarial perturbations
can significantly reduce CLIP's performance on new tasks. In this work, we
identify and explore the problem of \emph{adapting large-scale models for
zero-shot adversarial robustness}. We first identify two key factors during
model adaption -- training losses and adaptation methods -- that affect the
model's zero-shot adversarial robustness. We then propose a text-guided
contrastive adversarial training loss, which aligns the text embeddings and the
adversarial visual features with contrastive learning on a small set of
training data. We apply this training loss to two adaption methods, model
finetuning and visual prompt tuning. We find that visual prompt tuning is more
effective in the absence of texts, while finetuning wins in the existence of
text guidance. Overall, our approach significantly improves the zero-shot
adversarial robustness over CLIP, seeing an average improvement of over 31
points over ImageNet and 15 zero-shot datasets. We hope this work can shed
light on understanding the zero-shot adversarial robustness of large-scale
models.
| no_new_dataset | 0.709799 |
2302.01740 | Gorka Abad | Gorka Abad, Jing Xu, Stefanos Koffas, Behrad Tajalli, Stjepan Picek,
Mauro Conti | SoK: A Systematic Evaluation of Backdoor Trigger Characteristics in
Image Classification | null | null | null | null | cs.CV cs.CR cs.LG | http://creativecommons.org/licenses/by/4.0/ | Deep learning achieves outstanding results in many machine learning tasks.
Nevertheless, it is vulnerable to backdoor attacks that modify the training set
to embed a secret functionality in the trained model. The modified training
samples have a secret property, i. e., a trigger. At inference time, the secret
functionality is activated when the input contains the trigger, while the model
functions correctly in other cases. While there are many known backdoor attacks
(and defenses), deploying a stealthy attack is still far from trivial.
Successfully creating backdoor triggers depends on numerous parameters.
Unfortunately, research has not yet determined which parameters contribute most
to the attack performance.
This paper systematically analyzes the most relevant parameters for the
backdoor attacks, i.e., trigger size, position, color, and poisoning rate.
Using transfer learning, which is very common in computer vision, we evaluate
the attack on state-of-the-art models (ResNet, VGG, AlexNet, and GoogLeNet) and
datasets (MNIST, CIFAR10, and TinyImageNet). Our attacks cover the majority of
backdoor settings in research, providing concrete directions for future works.
Our code is publicly available to facilitate the reproducibility of our
results.
| [
{
"version": "v1",
"created": "Fri, 3 Feb 2023 14:00:05 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 09:01:24 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Abad",
"Gorka",
""
],
[
"Xu",
"Jing",
""
],
[
"Koffas",
"Stefanos",
""
],
[
"Tajalli",
"Behrad",
""
],
[
"Picek",
"Stjepan",
""
],
[
"Conti",
"Mauro",
""
]
] | TITLE: SoK: A Systematic Evaluation of Backdoor Trigger Characteristics in
Image Classification
ABSTRACT: Deep learning achieves outstanding results in many machine learning tasks.
Nevertheless, it is vulnerable to backdoor attacks that modify the training set
to embed a secret functionality in the trained model. The modified training
samples have a secret property, i. e., a trigger. At inference time, the secret
functionality is activated when the input contains the trigger, while the model
functions correctly in other cases. While there are many known backdoor attacks
(and defenses), deploying a stealthy attack is still far from trivial.
Successfully creating backdoor triggers depends on numerous parameters.
Unfortunately, research has not yet determined which parameters contribute most
to the attack performance.
This paper systematically analyzes the most relevant parameters for the
backdoor attacks, i.e., trigger size, position, color, and poisoning rate.
Using transfer learning, which is very common in computer vision, we evaluate
the attack on state-of-the-art models (ResNet, VGG, AlexNet, and GoogLeNet) and
datasets (MNIST, CIFAR10, and TinyImageNet). Our attacks cover the majority of
backdoor settings in research, providing concrete directions for future works.
Our code is publicly available to facilitate the reproducibility of our
results.
| no_new_dataset | 0.706577 |
2303.02297 | Yujie Lin | Yujie Lin, Chenyang Wang, Zhumin Chen, Zhaochun Ren, Xin Xin, Qiang
Yan, Maarten de Rijke, Xiuzhen Cheng, Pengjie Ren | A Self-Correcting Sequential Recommender | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sequential recommendations aim to capture users' preferences from their
historical interactions so as to predict the next item that they will interact
with. Sequential recommendation methods usually assume that all items in a
user's historical interactions reflect her/his preferences and transition
patterns between items. However, real-world interaction data is imperfect in
that (i) users might erroneously click on items, i.e., so-called misclicks on
irrelevant items, and (ii) users might miss items, i.e., unexposed relevant
items due to inaccurate recommendations. To tackle the two issues listed above,
we propose STEAM, a Self-correcTing sEquentiAl recoMmender. STEAM first
corrects an input item sequence by adjusting the misclicked and/or missed
items. It then uses the corrected item sequence to train a recommender and make
the next item prediction.We design an item-wise corrector that can adaptively
select one type of operation for each item in the sequence. The operation types
are 'keep', 'delete' and 'insert.' In order to train the item-wise corrector
without requiring additional labeling, we design two self-supervised learning
mechanisms: (i) deletion correction (i.e., deleting randomly inserted items),
and (ii) insertion correction (i.e., predicting randomly deleted items). We
integrate the corrector with the recommender by sharing the encoder and by
training them jointly. We conduct extensive experiments on three real-world
datasets and the experimental results demonstrate that STEAM outperforms
state-of-the-art sequential recommendation baselines. Our in-depth analyses
confirm that STEAM benefits from learning to correct the raw item sequences.
| [
{
"version": "v1",
"created": "Sat, 4 Mar 2023 02:35:29 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 03:02:41 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Lin",
"Yujie",
""
],
[
"Wang",
"Chenyang",
""
],
[
"Chen",
"Zhumin",
""
],
[
"Ren",
"Zhaochun",
""
],
[
"Xin",
"Xin",
""
],
[
"Yan",
"Qiang",
""
],
[
"de Rijke",
"Maarten",
""
],
[
"Cheng",
"Xiuzhen",
""
],
[
"Ren",
"Pengjie",
""
]
] | TITLE: A Self-Correcting Sequential Recommender
ABSTRACT: Sequential recommendations aim to capture users' preferences from their
historical interactions so as to predict the next item that they will interact
with. Sequential recommendation methods usually assume that all items in a
user's historical interactions reflect her/his preferences and transition
patterns between items. However, real-world interaction data is imperfect in
that (i) users might erroneously click on items, i.e., so-called misclicks on
irrelevant items, and (ii) users might miss items, i.e., unexposed relevant
items due to inaccurate recommendations. To tackle the two issues listed above,
we propose STEAM, a Self-correcTing sEquentiAl recoMmender. STEAM first
corrects an input item sequence by adjusting the misclicked and/or missed
items. It then uses the corrected item sequence to train a recommender and make
the next item prediction.We design an item-wise corrector that can adaptively
select one type of operation for each item in the sequence. The operation types
are 'keep', 'delete' and 'insert.' In order to train the item-wise corrector
without requiring additional labeling, we design two self-supervised learning
mechanisms: (i) deletion correction (i.e., deleting randomly inserted items),
and (ii) insertion correction (i.e., predicting randomly deleted items). We
integrate the corrector with the recommender by sharing the encoder and by
training them jointly. We conduct extensive experiments on three real-world
datasets and the experimental results demonstrate that STEAM outperforms
state-of-the-art sequential recommendation baselines. Our in-depth analyses
confirm that STEAM benefits from learning to correct the raw item sequences.
| no_new_dataset | 0.709372 |
2303.05275 | Davide Alessandro Coccomini | Davide Alessandro Coccomini, Andrea Esuli, Fabrizio Falchi, Claudio
Gennaro, Giuseppe Amato | Detecting Images Generated by Diffusers | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper explores the task of detecting images generated by text-to-image
diffusion models. To evaluate this, we consider images generated from captions
in the MSCOCO and Wikimedia datasets using two state-of-the-art models: Stable
Diffusion and GLIDE. Our experiments show that it is possible to detect the
generated images using simple Multi-Layer Perceptrons (MLPs), starting from
features extracted by CLIP, or traditional Convolutional Neural Networks
(CNNs). We also observe that models trained on images generated by Stable
Diffusion can detect images generated by GLIDE relatively well, however, the
reverse is not true. Lastly, we find that incorporating the associated textual
information with the images rarely leads to significant improvement in
detection results but that the type of subject depicted in the image can have a
significant impact on performance. This work provides insights into the
feasibility of detecting generated images, and has implications for security
and privacy concerns in real-world applications. The code to reproduce our
results is available at:
https://github.com/davide-coccomini/Detecting-Images-Generated-by-Diffusers
| [
{
"version": "v1",
"created": "Thu, 9 Mar 2023 14:14:29 GMT"
},
{
"version": "v2",
"created": "Tue, 4 Apr 2023 09:30:03 GMT"
},
{
"version": "v3",
"created": "Fri, 21 Apr 2023 14:17:10 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Coccomini",
"Davide Alessandro",
""
],
[
"Esuli",
"Andrea",
""
],
[
"Falchi",
"Fabrizio",
""
],
[
"Gennaro",
"Claudio",
""
],
[
"Amato",
"Giuseppe",
""
]
] | TITLE: Detecting Images Generated by Diffusers
ABSTRACT: This paper explores the task of detecting images generated by text-to-image
diffusion models. To evaluate this, we consider images generated from captions
in the MSCOCO and Wikimedia datasets using two state-of-the-art models: Stable
Diffusion and GLIDE. Our experiments show that it is possible to detect the
generated images using simple Multi-Layer Perceptrons (MLPs), starting from
features extracted by CLIP, or traditional Convolutional Neural Networks
(CNNs). We also observe that models trained on images generated by Stable
Diffusion can detect images generated by GLIDE relatively well, however, the
reverse is not true. Lastly, we find that incorporating the associated textual
information with the images rarely leads to significant improvement in
detection results but that the type of subject depicted in the image can have a
significant impact on performance. This work provides insights into the
feasibility of detecting generated images, and has implications for security
and privacy concerns in real-world applications. The code to reproduce our
results is available at:
https://github.com/davide-coccomini/Detecting-Images-Generated-by-Diffusers
| no_new_dataset | 0.712182 |
2303.07195 | Fran\c{c}ois Gauthier-Clerc | Fran\c{c}ois Gauthier-Clerc, Hoel Le Capitaine, Fabien Claveau,
Philippe Chevrel | Operating data of a specific Aquatic Center as a Benchmark for dynamic
model learning: search for a valid prediction model over an 8-hour horizon | null | null | null | null | cs.LG cs.SY eess.SY | http://creativecommons.org/licenses/by-sa/4.0/ | This article presents an identification benchmark based on data from a public
swimming pool in operation. Such a system is both a complex process and easily
understandable by all with regard to the stakes. Ultimately, the objective is
to reduce the energy bill while maintaining the level of quality of service.
This objective is general in scope and is not limited to public swimming pools.
This can be done effectively through what is known as economic predictive
control. This type of advanced control is based on a process model. It is the
aim of this article and the considered benchmark to show that such a dynamic
model can be obtained from operating data. For this, operational data is
formatted and shared, and model quality indicators are proposed. On this basis,
the first identification results illustrate the results obtained by a linear
multivariable model on the one hand, and by a neural dynamic model on the other
hand. The benchmark calls for other proposals and results from control and data
scientists for comparison.
| [
{
"version": "v1",
"created": "Fri, 10 Mar 2023 13:41:36 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 12:55:13 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Gauthier-Clerc",
"François",
""
],
[
"Capitaine",
"Hoel Le",
""
],
[
"Claveau",
"Fabien",
""
],
[
"Chevrel",
"Philippe",
""
]
] | TITLE: Operating data of a specific Aquatic Center as a Benchmark for dynamic
model learning: search for a valid prediction model over an 8-hour horizon
ABSTRACT: This article presents an identification benchmark based on data from a public
swimming pool in operation. Such a system is both a complex process and easily
understandable by all with regard to the stakes. Ultimately, the objective is
to reduce the energy bill while maintaining the level of quality of service.
This objective is general in scope and is not limited to public swimming pools.
This can be done effectively through what is known as economic predictive
control. This type of advanced control is based on a process model. It is the
aim of this article and the considered benchmark to show that such a dynamic
model can be obtained from operating data. For this, operational data is
formatted and shared, and model quality indicators are proposed. On this basis,
the first identification results illustrate the results obtained by a linear
multivariable model on the one hand, and by a neural dynamic model on the other
hand. The benchmark calls for other proposals and results from control and data
scientists for comparison.
| no_new_dataset | 0.705798 |
2303.16580 | Shenyuan Gao | Shenyuan Gao, Chunluan Zhou, Jun Zhang | Generalized Relation Modeling for Transformer Tracking | Accepted by CVPR 2023. v3: fix a typo in equation (7). Code and
models are publicly available at https://github.com/Little-Podi/GRM | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Compared with previous two-stream trackers, the recent one-stream tracking
pipeline, which allows earlier interaction between the template and search
region, has achieved a remarkable performance gain. However, existing
one-stream trackers always let the template interact with all parts inside the
search region throughout all the encoder layers. This could potentially lead to
target-background confusion when the extracted feature representations are not
sufficiently discriminative. To alleviate this issue, we propose a generalized
relation modeling method based on adaptive token division. The proposed method
is a generalized formulation of attention-based relation modeling for
Transformer tracking, which inherits the merits of both previous two-stream and
one-stream pipelines whilst enabling more flexible relation modeling by
selecting appropriate search tokens to interact with template tokens. An
attention masking strategy and the Gumbel-Softmax technique are introduced to
facilitate the parallel computation and end-to-end learning of the token
division module. Extensive experiments show that our method is superior to the
two-stream and one-stream pipelines and achieves state-of-the-art performance
on six challenging benchmarks with a real-time running speed.
| [
{
"version": "v1",
"created": "Wed, 29 Mar 2023 10:29:25 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Apr 2023 17:23:53 GMT"
},
{
"version": "v3",
"created": "Fri, 21 Apr 2023 14:26:50 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Gao",
"Shenyuan",
""
],
[
"Zhou",
"Chunluan",
""
],
[
"Zhang",
"Jun",
""
]
] | TITLE: Generalized Relation Modeling for Transformer Tracking
ABSTRACT: Compared with previous two-stream trackers, the recent one-stream tracking
pipeline, which allows earlier interaction between the template and search
region, has achieved a remarkable performance gain. However, existing
one-stream trackers always let the template interact with all parts inside the
search region throughout all the encoder layers. This could potentially lead to
target-background confusion when the extracted feature representations are not
sufficiently discriminative. To alleviate this issue, we propose a generalized
relation modeling method based on adaptive token division. The proposed method
is a generalized formulation of attention-based relation modeling for
Transformer tracking, which inherits the merits of both previous two-stream and
one-stream pipelines whilst enabling more flexible relation modeling by
selecting appropriate search tokens to interact with template tokens. An
attention masking strategy and the Gumbel-Softmax technique are introduced to
facilitate the parallel computation and end-to-end learning of the token
division module. Extensive experiments show that our method is superior to the
two-stream and one-stream pipelines and achieves state-of-the-art performance
on six challenging benchmarks with a real-time running speed.
| no_new_dataset | 0.709799 |
2304.05243 | Klaudia Balazy | Klaudia Ba{\l}azy, {\L}ukasz Struski, Marek \'Smieja, Jacek Tabor | r-softmax: Generalized Softmax with Controllable Sparsity Rate | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Nowadays artificial neural network models achieve remarkable results in many
disciplines. Functions mapping the representation provided by the model to the
probability distribution are the inseparable aspect of deep learning solutions.
Although softmax is a commonly accepted probability mapping function in the
machine learning community, it cannot return sparse outputs and always spreads
the positive probability to all positions. In this paper, we propose r-softmax,
a modification of the softmax, outputting sparse probability distribution with
controllable sparsity rate. In contrast to the existing sparse probability
mapping functions, we provide an intuitive mechanism for controlling the output
sparsity level. We show on several multi-label datasets that r-softmax
outperforms other sparse alternatives to softmax and is highly competitive with
the original softmax. We also apply r-softmax to the self-attention module of a
pre-trained transformer language model and demonstrate that it leads to
improved performance when fine-tuning the model on different natural language
processing tasks.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 14:28:29 GMT"
},
{
"version": "v2",
"created": "Wed, 12 Apr 2023 16:30:16 GMT"
},
{
"version": "v3",
"created": "Fri, 21 Apr 2023 14:41:43 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Bałazy",
"Klaudia",
""
],
[
"Struski",
"Łukasz",
""
],
[
"Śmieja",
"Marek",
""
],
[
"Tabor",
"Jacek",
""
]
] | TITLE: r-softmax: Generalized Softmax with Controllable Sparsity Rate
ABSTRACT: Nowadays artificial neural network models achieve remarkable results in many
disciplines. Functions mapping the representation provided by the model to the
probability distribution are the inseparable aspect of deep learning solutions.
Although softmax is a commonly accepted probability mapping function in the
machine learning community, it cannot return sparse outputs and always spreads
the positive probability to all positions. In this paper, we propose r-softmax,
a modification of the softmax, outputting sparse probability distribution with
controllable sparsity rate. In contrast to the existing sparse probability
mapping functions, we provide an intuitive mechanism for controlling the output
sparsity level. We show on several multi-label datasets that r-softmax
outperforms other sparse alternatives to softmax and is highly competitive with
the original softmax. We also apply r-softmax to the self-attention module of a
pre-trained transformer language model and demonstrate that it leads to
improved performance when fine-tuning the model on different natural language
processing tasks.
| no_new_dataset | 0.709252 |
2304.07099 | Amit Shomer | Amit Shomer, Shai Avidan | Prior based Sampling for Adaptive LiDAR | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose SampleDepth, a Convolutional Neural Network (CNN), that is suited
for an adaptive LiDAR. Typically,LiDAR sampling strategy is pre-defined,
constant and independent of the observed scene. Instead of letting a LiDAR
sample the scene in this agnostic fashion, SampleDepth determines, adaptively,
where it is best to sample the current frame. To do that, SampleDepth uses
depth samples from previous time steps to predict a sampling mask for the
current frame. Crucially, SampleDepth is trained to optimize the performance of
a depth completion downstream task. SampleDepth is evaluated on two different
depth completion networks and two LiDAR datasets, KITTI Depth Completion and
the newly introduced synthetic dataset, SHIFT. We show that SampleDepth is
effective and suitable for different depth completion downstream tasks.
| [
{
"version": "v1",
"created": "Fri, 14 Apr 2023 12:41:00 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Shomer",
"Amit",
""
],
[
"Avidan",
"Shai",
""
]
] | TITLE: Prior based Sampling for Adaptive LiDAR
ABSTRACT: We propose SampleDepth, a Convolutional Neural Network (CNN), that is suited
for an adaptive LiDAR. Typically,LiDAR sampling strategy is pre-defined,
constant and independent of the observed scene. Instead of letting a LiDAR
sample the scene in this agnostic fashion, SampleDepth determines, adaptively,
where it is best to sample the current frame. To do that, SampleDepth uses
depth samples from previous time steps to predict a sampling mask for the
current frame. Crucially, SampleDepth is trained to optimize the performance of
a depth completion downstream task. SampleDepth is evaluated on two different
depth completion networks and two LiDAR datasets, KITTI Depth Completion and
the newly introduced synthetic dataset, SHIFT. We show that SampleDepth is
effective and suitable for different depth completion downstream tasks.
| new_dataset | 0.716609 |
2304.08838 | Tiantian Liu | Tiantian Liu, Huan Li, Hua Lu, Muhammad Aamir Cheema, Harry Kai-Ho
Chan | Contact Tracing over Uncertain Indoor Positioning Data (Extended
Version) | Accepted by TKDE (April.2023) | null | null | null | cs.DB | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Pandemics often cause dramatic losses of human lives and impact our societies
in many aspects such as public health, tourism, and economy. To contain the
spread of an epidemic like COVID-19, efficient and effective contact tracing is
important, especially in indoor venues where the risk of infection is higher.
In this work, we formulate and study a novel query called Indoor Contact Query
(ICQ) over raw, uncertain indoor positioning data that digitalizes people's
movements indoors. Given a query object o, e.g., a person confirmed to be a
virus carrier, an ICQ analyzes uncertain indoor positioning data to find
objects that most likely had close contact with o for a long period of time. To
process ICQ, we propose a set of techniques. First, we design an enhanced
indoor graph model to organize different types of data necessary for ICQ.
Second, for indoor moving objects, we devise methods to determine uncertain
regions and to derive positioning samples missing in the raw data. Third, we
propose a query processing framework with a close contact determination method,
a search algorithm, and the acceleration strategies. We conduct extensive
experiments on synthetic and real datasets to evaluate our proposals. The
results demonstrate the efficiency and effectiveness of our proposals.
| [
{
"version": "v1",
"created": "Tue, 18 Apr 2023 09:05:53 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 17:06:47 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Liu",
"Tiantian",
""
],
[
"Li",
"Huan",
""
],
[
"Lu",
"Hua",
""
],
[
"Cheema",
"Muhammad Aamir",
""
],
[
"Chan",
"Harry Kai-Ho",
""
]
] | TITLE: Contact Tracing over Uncertain Indoor Positioning Data (Extended
Version)
ABSTRACT: Pandemics often cause dramatic losses of human lives and impact our societies
in many aspects such as public health, tourism, and economy. To contain the
spread of an epidemic like COVID-19, efficient and effective contact tracing is
important, especially in indoor venues where the risk of infection is higher.
In this work, we formulate and study a novel query called Indoor Contact Query
(ICQ) over raw, uncertain indoor positioning data that digitalizes people's
movements indoors. Given a query object o, e.g., a person confirmed to be a
virus carrier, an ICQ analyzes uncertain indoor positioning data to find
objects that most likely had close contact with o for a long period of time. To
process ICQ, we propose a set of techniques. First, we design an enhanced
indoor graph model to organize different types of data necessary for ICQ.
Second, for indoor moving objects, we devise methods to determine uncertain
regions and to derive positioning samples missing in the raw data. Third, we
propose a query processing framework with a close contact determination method,
a search algorithm, and the acceleration strategies. We conduct extensive
experiments on synthetic and real datasets to evaluate our proposals. The
results demonstrate the efficiency and effectiveness of our proposals.
| no_new_dataset | 0.708389 |
2304.09938 | Claire Pagetti | M\'elanie Ducoffe, Maxime Carrere, L\'eo F\'eliers, Adrien Gauffriau,
Vincent Mussot, Claire Pagetti, Thierry Sammour | LARD -- Landing Approach Runway Detection -- Dataset for Vision Based
Landing | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As the interest in autonomous systems continues to grow, one of the major
challenges is collecting sufficient and representative real-world data. Despite
the strong practical and commercial interest in autonomous landing systems in
the aerospace field, there is a lack of open-source datasets of aerial images.
To address this issue, we present a dataset-lard-of high-quality aerial images
for the task of runway detection during approach and landing phases. Most of
the dataset is composed of synthetic images but we also provide manually
labelled images from real landing footages, to extend the detection task to a
more realistic setting. In addition, we offer the generator which can produce
such synthetic front-view images and enables automatic annotation of the runway
corners through geometric transformations. This dataset paves the way for
further research such as the analysis of dataset quality or the development of
models to cope with the detection tasks. Find data, code and more up-to-date
information at https://github.com/deel-ai/LARD
| [
{
"version": "v1",
"created": "Wed, 5 Apr 2023 08:25:55 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 13:58:29 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Ducoffe",
"Mélanie",
""
],
[
"Carrere",
"Maxime",
""
],
[
"Féliers",
"Léo",
""
],
[
"Gauffriau",
"Adrien",
""
],
[
"Mussot",
"Vincent",
""
],
[
"Pagetti",
"Claire",
""
],
[
"Sammour",
"Thierry",
""
]
] | TITLE: LARD -- Landing Approach Runway Detection -- Dataset for Vision Based
Landing
ABSTRACT: As the interest in autonomous systems continues to grow, one of the major
challenges is collecting sufficient and representative real-world data. Despite
the strong practical and commercial interest in autonomous landing systems in
the aerospace field, there is a lack of open-source datasets of aerial images.
To address this issue, we present a dataset-lard-of high-quality aerial images
for the task of runway detection during approach and landing phases. Most of
the dataset is composed of synthetic images but we also provide manually
labelled images from real landing footages, to extend the detection task to a
more realistic setting. In addition, we offer the generator which can produce
such synthetic front-view images and enables automatic annotation of the runway
corners through geometric transformations. This dataset paves the way for
further research such as the analysis of dataset quality or the development of
models to cope with the detection tasks. Find data, code and more up-to-date
information at https://github.com/deel-ai/LARD
| new_dataset | 0.714634 |
2304.10379 | Baptiste Roziere | Ruba Mutasim, Gabriel Synnaeve, David Pichardie, Baptiste Rozi\`ere | Leveraging Static Analysis for Bug Repair | 13 pages. DL4C 2023 | null | null | null | cs.SE | http://creativecommons.org/licenses/by-sa/4.0/ | We propose a method combining machine learning with a static analysis tool
(i.e. Infer) to automatically repair source code. Machine Learning methods
perform well for producing idiomatic source code. However, their output is
sometimes difficult to trust as language models can output incorrect code with
high confidence. Static analysis tools are trustable, but also less flexible
and produce non-idiomatic code. In this paper, we propose to fix resource leak
bugs in IR space, and to use a sequence-to-sequence model to propose fix in
source code space. We also study several decoding strategies, and use Infer to
filter the output of the model. On a dataset of CodeNet submissions with
potential resource leak bugs, our method is able to find a function with the
same semantics that does not raise a warning with around 97% precision and 66%
recall.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 15:20:08 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Apr 2023 14:50:44 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Mutasim",
"Ruba",
""
],
[
"Synnaeve",
"Gabriel",
""
],
[
"Pichardie",
"David",
""
],
[
"Rozière",
"Baptiste",
""
]
] | TITLE: Leveraging Static Analysis for Bug Repair
ABSTRACT: We propose a method combining machine learning with a static analysis tool
(i.e. Infer) to automatically repair source code. Machine Learning methods
perform well for producing idiomatic source code. However, their output is
sometimes difficult to trust as language models can output incorrect code with
high confidence. Static analysis tools are trustable, but also less flexible
and produce non-idiomatic code. In this paper, we propose to fix resource leak
bugs in IR space, and to use a sequence-to-sequence model to propose fix in
source code space. We also study several decoding strategies, and use Infer to
filter the output of the model. On a dataset of CodeNet submissions with
potential resource leak bugs, our method is able to find a function with the
same semantics that does not raise a warning with around 97% precision and 66%
recall.
| no_new_dataset | 0.709655 |
2304.10539 | Wenqiao Zhang | Wenqiao Zhang, Changshuo Liu, Lingze Zeng, Beng Chin Ooi, Siliang
Tang, Yueting Zhuang | Learning in Imperfect Environment: Multi-Label Classification with
Long-Tailed Distribution and Partial Labels | null | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Conventional multi-label classification (MLC) methods assume that all samples
are fully labeled and identically distributed. Unfortunately, this assumption
is unrealistic in large-scale MLC data that has long-tailed (LT) distribution
and partial labels (PL). To address the problem, we introduce a novel task,
Partial labeling and Long-Tailed Multi-Label Classification (PLT-MLC), to
jointly consider the above two imperfect learning environments. Not
surprisingly, we find that most LT-MLC and PL-MLC approaches fail to solve the
PLT-MLC, resulting in significant performance degradation on the two proposed
PLT-MLC benchmarks. Therefore, we propose an end-to-end learning framework:
\textbf{CO}rrection $\rightarrow$ \textbf{M}odificat\textbf{I}on $\rightarrow$
balan\textbf{C}e, abbreviated as \textbf{\method{}}. Our bootstrapping
philosophy is to simultaneously correct the missing labels (Correction) with
convinced prediction confidence over a class-aware threshold and to learn from
these recall labels during training. We next propose a novel multi-focal
modifier loss that simultaneously addresses head-tail imbalance and
positive-negative imbalance to adaptively modify the attention to different
samples (Modification) under the LT class distribution. In addition, we develop
a balanced training strategy by distilling the model's learning effect from
head and tail samples, and thus design a balanced classifier (Balance)
conditioned on the head and tail learning effect to maintain stable performance
for all samples. Our experimental study shows that the proposed \method{}
significantly outperforms general MLC, LT-MLC and PL-MLC methods in terms of
effectiveness and robustness on our newly created PLT-MLC datasets.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 20:05:08 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Zhang",
"Wenqiao",
""
],
[
"Liu",
"Changshuo",
""
],
[
"Zeng",
"Lingze",
""
],
[
"Ooi",
"Beng Chin",
""
],
[
"Tang",
"Siliang",
""
],
[
"Zhuang",
"Yueting",
""
]
] | TITLE: Learning in Imperfect Environment: Multi-Label Classification with
Long-Tailed Distribution and Partial Labels
ABSTRACT: Conventional multi-label classification (MLC) methods assume that all samples
are fully labeled and identically distributed. Unfortunately, this assumption
is unrealistic in large-scale MLC data that has long-tailed (LT) distribution
and partial labels (PL). To address the problem, we introduce a novel task,
Partial labeling and Long-Tailed Multi-Label Classification (PLT-MLC), to
jointly consider the above two imperfect learning environments. Not
surprisingly, we find that most LT-MLC and PL-MLC approaches fail to solve the
PLT-MLC, resulting in significant performance degradation on the two proposed
PLT-MLC benchmarks. Therefore, we propose an end-to-end learning framework:
\textbf{CO}rrection $\rightarrow$ \textbf{M}odificat\textbf{I}on $\rightarrow$
balan\textbf{C}e, abbreviated as \textbf{\method{}}. Our bootstrapping
philosophy is to simultaneously correct the missing labels (Correction) with
convinced prediction confidence over a class-aware threshold and to learn from
these recall labels during training. We next propose a novel multi-focal
modifier loss that simultaneously addresses head-tail imbalance and
positive-negative imbalance to adaptively modify the attention to different
samples (Modification) under the LT class distribution. In addition, we develop
a balanced training strategy by distilling the model's learning effect from
head and tail samples, and thus design a balanced classifier (Balance)
conditioned on the head and tail learning effect to maintain stable performance
for all samples. Our experimental study shows that the proposed \method{}
significantly outperforms general MLC, LT-MLC and PL-MLC methods in terms of
effectiveness and robustness on our newly created PLT-MLC datasets.
| no_new_dataset | 0.711049 |
2304.10548 | Ziang Xiao | Ziang Xiao, Xingdi Yuan, Q. Vera Liao, Rania Abdelghani, Pierre-Yves
Oudeyer | Supporting Qualitative Analysis with Large Language Models: Combining
Codebook with GPT-3 for Deductive Coding | 28th International Conference on Intelligent User Interfaces (IUI '23
Companion), March 27--31, 2023, Sydney, NSW, Australia | null | 10.1145/3581754.3584136 | null | cs.CL cs.AI cs.HC | http://creativecommons.org/licenses/by/4.0/ | Qualitative analysis of textual contents unpacks rich and valuable
information by assigning labels to the data. However, this process is often
labor-intensive, particularly when working with large datasets. While recent
AI-based tools demonstrate utility, researchers may not have readily available
AI resources and expertise, let alone be challenged by the limited
generalizability of those task-specific models. In this study, we explored the
use of large language models (LLMs) in supporting deductive coding, a major
category of qualitative analysis where researchers use pre-determined codebooks
to label the data into a fixed set of codes. Instead of training task-specific
models, a pre-trained LLM could be used directly for various tasks without
fine-tuning through prompt learning. Using a curiosity-driven questions coding
task as a case study, we found, by combining GPT-3 with expert-drafted
codebooks, our proposed approach achieved fair to substantial agreements with
expert-coded results. We lay out challenges and opportunities in using LLMs to
support qualitative coding and beyond.
| [
{
"version": "v1",
"created": "Mon, 17 Apr 2023 04:52:43 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Xiao",
"Ziang",
""
],
[
"Yuan",
"Xingdi",
""
],
[
"Liao",
"Q. Vera",
""
],
[
"Abdelghani",
"Rania",
""
],
[
"Oudeyer",
"Pierre-Yves",
""
]
] | TITLE: Supporting Qualitative Analysis with Large Language Models: Combining
Codebook with GPT-3 for Deductive Coding
ABSTRACT: Qualitative analysis of textual contents unpacks rich and valuable
information by assigning labels to the data. However, this process is often
labor-intensive, particularly when working with large datasets. While recent
AI-based tools demonstrate utility, researchers may not have readily available
AI resources and expertise, let alone be challenged by the limited
generalizability of those task-specific models. In this study, we explored the
use of large language models (LLMs) in supporting deductive coding, a major
category of qualitative analysis where researchers use pre-determined codebooks
to label the data into a fixed set of codes. Instead of training task-specific
models, a pre-trained LLM could be used directly for various tasks without
fine-tuning through prompt learning. Using a curiosity-driven questions coding
task as a case study, we found, by combining GPT-3 with expert-drafted
codebooks, our proposed approach achieved fair to substantial agreements with
expert-coded results. We lay out challenges and opportunities in using LLMs to
support qualitative coding and beyond.
| no_new_dataset | 0.709585 |
2304.10551 | Qianhui Sun | Qianhui Sun, Qingyu Yang, Chongyi Li, Shangchen Zhou, Ruicheng Feng,
Yuekun Dai, Wenxiu Sun, Qingpeng Zhu, Chen Change Loy, Jinwei Gu | MIPI 2023 Challenge on RGBW Remosaic: Methods and Results | CVPR 2023 Mobile Intelligent Photography and Imaging (MIPI)
Workshop--RGBW Sensor Remosaic Challenge Report. Website:
https://mipi-challenge.org/MIPI2023/. arXiv admin note: substantial text
overlap with arXiv:2209.08471, arXiv:2209.07060, arXiv:2209.07530,
arXiv:2304.10089 | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Developing and integrating advanced image sensors with novel algorithms in
camera systems are prevalent with the increasing demand for computational
photography and imaging on mobile platforms. However, the lack of high-quality
data for research and the rare opportunity for an in-depth exchange of views
from industry and academia constrain the development of mobile intelligent
photography and imaging (MIPI). With the success of the 1st MIPI Workshop@ECCV
2022, we introduce the second MIPI challenge, including four tracks focusing on
novel image sensors and imaging algorithms. This paper summarizes and reviews
the RGBW Joint Remosaic and Denoise track on MIPI 2023. In total, 81
participants were successfully registered, and 4 teams submitted results in the
final testing phase. The final results are evaluated using objective metrics,
including PSNR, SSIM, LPIPS, and KLD. A detailed description of the top three
models developed in this challenge is provided in this paper. More details of
this challenge and the link to the dataset can be found at
https://mipi-challenge.org/MIPI2023/.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 05:12:42 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Sun",
"Qianhui",
""
],
[
"Yang",
"Qingyu",
""
],
[
"Li",
"Chongyi",
""
],
[
"Zhou",
"Shangchen",
""
],
[
"Feng",
"Ruicheng",
""
],
[
"Dai",
"Yuekun",
""
],
[
"Sun",
"Wenxiu",
""
],
[
"Zhu",
"Qingpeng",
""
],
[
"Loy",
"Chen Change",
""
],
[
"Gu",
"Jinwei",
""
]
] | TITLE: MIPI 2023 Challenge on RGBW Remosaic: Methods and Results
ABSTRACT: Developing and integrating advanced image sensors with novel algorithms in
camera systems are prevalent with the increasing demand for computational
photography and imaging on mobile platforms. However, the lack of high-quality
data for research and the rare opportunity for an in-depth exchange of views
from industry and academia constrain the development of mobile intelligent
photography and imaging (MIPI). With the success of the 1st MIPI Workshop@ECCV
2022, we introduce the second MIPI challenge, including four tracks focusing on
novel image sensors and imaging algorithms. This paper summarizes and reviews
the RGBW Joint Remosaic and Denoise track on MIPI 2023. In total, 81
participants were successfully registered, and 4 teams submitted results in the
final testing phase. The final results are evaluated using objective metrics,
including PSNR, SSIM, LPIPS, and KLD. A detailed description of the top three
models developed in this challenge is provided in this paper. More details of
this challenge and the link to the dataset can be found at
https://mipi-challenge.org/MIPI2023/.
| no_new_dataset | 0.700447 |
2304.10572 | Pranay Mundra | Pranay Mundra, Jianhao Zhang, Fatemeh Nargesian, Nikolaus Augsten | KOIOS: Top-k Semantic Overlap Set Search | null | null | null | null | cs.DB | http://creativecommons.org/licenses/by/4.0/ | We study the top-k set similarity search problem using semantic overlap.
While vanilla overlap requires exact matches between set elements, semantic
overlap allows elements that are syntactically different but semantically
related to increase the overlap. The semantic overlap is the maximum matching
score of a bipartite graph, where an edge weight between two set elements is
defined by a user-defined similarity function, e.g., cosine similarity between
embeddings. Common techniques like token indexes fail for semantic search since
similar elements may be unrelated at the character level. Further, verifying
candidates is expensive (cubic versus linear for syntactic overlap), calling
for highly selective filters. We propose KOIOS, the first exact and efficient
algorithm for semantic overlap search. KOIOS leverages sophisticated filters to
minimize the number of required graph-matching calculations. Our experiments
show that for medium to large sets less than 5% of the candidate sets need
verification, and more than half of those sets are further pruned without
requiring the expensive graph matching. We show the efficiency of our algorithm
on four real datasets and demonstrate the improved result quality of semantic
over vanilla set similarity search.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 18:03:21 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Mundra",
"Pranay",
""
],
[
"Zhang",
"Jianhao",
""
],
[
"Nargesian",
"Fatemeh",
""
],
[
"Augsten",
"Nikolaus",
""
]
] | TITLE: KOIOS: Top-k Semantic Overlap Set Search
ABSTRACT: We study the top-k set similarity search problem using semantic overlap.
While vanilla overlap requires exact matches between set elements, semantic
overlap allows elements that are syntactically different but semantically
related to increase the overlap. The semantic overlap is the maximum matching
score of a bipartite graph, where an edge weight between two set elements is
defined by a user-defined similarity function, e.g., cosine similarity between
embeddings. Common techniques like token indexes fail for semantic search since
similar elements may be unrelated at the character level. Further, verifying
candidates is expensive (cubic versus linear for syntactic overlap), calling
for highly selective filters. We propose KOIOS, the first exact and efficient
algorithm for semantic overlap search. KOIOS leverages sophisticated filters to
minimize the number of required graph-matching calculations. Our experiments
show that for medium to large sets less than 5% of the candidate sets need
verification, and more than half of those sets are further pruned without
requiring the expensive graph matching. We show the efficiency of our algorithm
on four real datasets and demonstrate the improved result quality of semantic
over vanilla set similarity search.
| no_new_dataset | 0.711656 |
2304.10588 | Zhuangzhuang Dai | Zhuangzhuang Dai, Jinha Park, Aleksandra Kaszowska, Chen Li | Detecting Worker Attention Lapses in Human-Robot Interaction: An Eye
Tracking and Multimodal Sensing Study | 6 pages | null | null | null | cs.RO cs.HC | http://creativecommons.org/licenses/by/4.0/ | The advent of industrial robotics and autonomous systems endow human-robot
collaboration in a massive scale. However, current industrial robots are
restrained in co-working with human in close proximity due to inability of
interpreting human agents' attention. Human attention study is non-trivial
since it involves multiple aspects of the mind: perception, memory, problem
solving, and consciousness. Human attention lapses are particularly problematic
and potentially catastrophic in industrial workplace, from assembling
electronics to operating machines. Attention is indeed complex and cannot be
easily measured with single-modality sensors. Eye state, head pose, posture,
and manifold environment stimulus could all play a part in attention lapses. To
this end, we propose a pipeline to annotate multimodal dataset of human
attention tracking, including eye tracking, fixation detection, third-person
surveillance camera, and sound. We produce a pilot dataset containing two fully
annotated phone assembly sequences in a realistic manufacturing environment. We
evaluate existing fatigue and drowsiness prediction methods for attention lapse
detection. Experimental results show that human attention lapses in production
scenarios are more subtle and imperceptible than well-studied fatigue and
drowsiness.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 18:20:26 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Dai",
"Zhuangzhuang",
""
],
[
"Park",
"Jinha",
""
],
[
"Kaszowska",
"Aleksandra",
""
],
[
"Li",
"Chen",
""
]
] | TITLE: Detecting Worker Attention Lapses in Human-Robot Interaction: An Eye
Tracking and Multimodal Sensing Study
ABSTRACT: The advent of industrial robotics and autonomous systems endow human-robot
collaboration in a massive scale. However, current industrial robots are
restrained in co-working with human in close proximity due to inability of
interpreting human agents' attention. Human attention study is non-trivial
since it involves multiple aspects of the mind: perception, memory, problem
solving, and consciousness. Human attention lapses are particularly problematic
and potentially catastrophic in industrial workplace, from assembling
electronics to operating machines. Attention is indeed complex and cannot be
easily measured with single-modality sensors. Eye state, head pose, posture,
and manifold environment stimulus could all play a part in attention lapses. To
this end, we propose a pipeline to annotate multimodal dataset of human
attention tracking, including eye tracking, fixation detection, third-person
surveillance camera, and sound. We produce a pilot dataset containing two fully
annotated phone assembly sequences in a realistic manufacturing environment. We
evaluate existing fatigue and drowsiness prediction methods for attention lapse
detection. Experimental results show that human attention lapses in production
scenarios are more subtle and imperceptible than well-studied fatigue and
drowsiness.
| new_dataset | 0.715064 |
2304.10616 | Duc Hoa Tran | Duc Hoa Tran, Michel Meunier, Farida Cheriet | Multi-domain learning CNN model for microscopy image classification | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For any type of microscopy image, getting a deep learning model to work well
requires considerable effort to select a suitable architecture and time to
train it. As there is a wide range of microscopes and experimental setups,
designing a single model that can apply to multiple imaging domains, instead of
having multiple per-domain models, becomes more essential. This task is
challenging and somehow overlooked in the literature. In this paper, we present
a multi-domain learning architecture for the classification of microscopy
images that differ significantly in types and contents. Unlike previous methods
that are computationally intensive, we have developed a compact model, called
Mobincep, by combining the simple but effective techniques of depth-wise
separable convolution and the inception module. We also introduce a new
optimization technique to regulate the latent feature space during training to
improve the network's performance. We evaluated our model on three different
public datasets and compared its performance in single-domain and
multiple-domain learning modes. The proposed classifier surpasses
state-of-the-art results and is robust for limited labeled data. Moreover, it
helps to eliminate the burden of designing a new network when switching to new
experiments.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 19:32:23 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Tran",
"Duc Hoa",
""
],
[
"Meunier",
"Michel",
""
],
[
"Cheriet",
"Farida",
""
]
] | TITLE: Multi-domain learning CNN model for microscopy image classification
ABSTRACT: For any type of microscopy image, getting a deep learning model to work well
requires considerable effort to select a suitable architecture and time to
train it. As there is a wide range of microscopes and experimental setups,
designing a single model that can apply to multiple imaging domains, instead of
having multiple per-domain models, becomes more essential. This task is
challenging and somehow overlooked in the literature. In this paper, we present
a multi-domain learning architecture for the classification of microscopy
images that differ significantly in types and contents. Unlike previous methods
that are computationally intensive, we have developed a compact model, called
Mobincep, by combining the simple but effective techniques of depth-wise
separable convolution and the inception module. We also introduce a new
optimization technique to regulate the latent feature space during training to
improve the network's performance. We evaluated our model on three different
public datasets and compared its performance in single-domain and
multiple-domain learning modes. The proposed classifier surpasses
state-of-the-art results and is robust for limited labeled data. Moreover, it
helps to eliminate the burden of designing a new network when switching to new
experiments.
| no_new_dataset | 0.710208 |
2304.10618 | Zachary Susskind | Zachary Susskind, Aman Arora, Igor D. S. Miranda, Alan T. L. Bacellar,
Luis A. Q. Villon, Rafael F. Katopodis, Leandro S. de Araujo, Diego L. C.
Dutra, Priscila M. V. Lima, Felipe M. G. Franca, Mauricio Breternitz Jr., and
Lizy K. John | ULEEN: A Novel Architecture for Ultra Low-Energy Edge Neural Networks | 14 pages, 14 figures Portions of this article draw heavily from
arXiv:2203.01479, most notably sections 5E and 5F.2 | null | null | null | cs.AR eess.SP | http://creativecommons.org/licenses/by/4.0/ | The deployment of AI models on low-power, real-time edge devices requires
accelerators for which energy, latency, and area are all first-order concerns.
There are many approaches to enabling deep neural networks (DNNs) in this
domain, including pruning, quantization, compression, and binary neural
networks (BNNs), but with the emergence of the "extreme edge", there is now a
demand for even more efficient models. In order to meet the constraints of
ultra-low-energy devices, we propose ULEEN, a model architecture based on
weightless neural networks. Weightless neural networks (WNNs) are a class of
neural model which use table lookups, not arithmetic, to perform computation.
The elimination of energy-intensive arithmetic operations makes WNNs
theoretically well suited for edge inference; however, they have historically
suffered from poor accuracy and excessive memory usage. ULEEN incorporates
algorithmic improvements and a novel training strategy inspired by BNNs to make
significant strides in improving accuracy and reducing model size. We compare
FPGA and ASIC implementations of an inference accelerator for ULEEN against
edge-optimized DNN and BNN devices. On a Xilinx Zynq Z-7045 FPGA, we
demonstrate classification on the MNIST dataset at 14.3 million inferences per
second (13 million inferences/Joule) with 0.21 $\mu$s latency and 96.2%
accuracy, while Xilinx FINN achieves 12.3 million inferences per second (1.69
million inferences/Joule) with 0.31 $\mu$s latency and 95.83% accuracy. In a
45nm ASIC, we achieve 5.1 million inferences/Joule and 38.5 million
inferences/second at 98.46% accuracy, while a quantized Bit Fusion model
achieves 9230 inferences/Joule and 19,100 inferences/second at 99.35% accuracy.
In our search for ever more efficient edge devices, ULEEN shows that WNNs are
deserving of consideration.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 19:40:01 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Susskind",
"Zachary",
""
],
[
"Arora",
"Aman",
""
],
[
"Miranda",
"Igor D. S.",
""
],
[
"Bacellar",
"Alan T. L.",
""
],
[
"Villon",
"Luis A. Q.",
""
],
[
"Katopodis",
"Rafael F.",
""
],
[
"de Araujo",
"Leandro S.",
""
],
[
"Dutra",
"Diego L. C.",
""
],
[
"Lima",
"Priscila M. V.",
""
],
[
"Franca",
"Felipe M. G.",
""
],
[
"Breternitz",
"Mauricio",
"Jr."
],
[
"John",
"Lizy K.",
""
]
] | TITLE: ULEEN: A Novel Architecture for Ultra Low-Energy Edge Neural Networks
ABSTRACT: The deployment of AI models on low-power, real-time edge devices requires
accelerators for which energy, latency, and area are all first-order concerns.
There are many approaches to enabling deep neural networks (DNNs) in this
domain, including pruning, quantization, compression, and binary neural
networks (BNNs), but with the emergence of the "extreme edge", there is now a
demand for even more efficient models. In order to meet the constraints of
ultra-low-energy devices, we propose ULEEN, a model architecture based on
weightless neural networks. Weightless neural networks (WNNs) are a class of
neural model which use table lookups, not arithmetic, to perform computation.
The elimination of energy-intensive arithmetic operations makes WNNs
theoretically well suited for edge inference; however, they have historically
suffered from poor accuracy and excessive memory usage. ULEEN incorporates
algorithmic improvements and a novel training strategy inspired by BNNs to make
significant strides in improving accuracy and reducing model size. We compare
FPGA and ASIC implementations of an inference accelerator for ULEEN against
edge-optimized DNN and BNN devices. On a Xilinx Zynq Z-7045 FPGA, we
demonstrate classification on the MNIST dataset at 14.3 million inferences per
second (13 million inferences/Joule) with 0.21 $\mu$s latency and 96.2%
accuracy, while Xilinx FINN achieves 12.3 million inferences per second (1.69
million inferences/Joule) with 0.31 $\mu$s latency and 95.83% accuracy. In a
45nm ASIC, we achieve 5.1 million inferences/Joule and 38.5 million
inferences/second at 98.46% accuracy, while a quantized Bit Fusion model
achieves 9230 inferences/Joule and 19,100 inferences/second at 99.35% accuracy.
In our search for ever more efficient edge devices, ULEEN shows that WNNs are
deserving of consideration.
| no_new_dataset | 0.710785 |
2304.10622 | Nilantha Premakumara Weerappulige | Nilantha Premakumara, Brian Jalaian, Niranjan Suri and Hooman Samani | Enhancing object detection robustness: A synthetic and natural
perturbation approach | 09 pages, 4 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Robustness against real-world distribution shifts is crucial for the
successful deployment of object detection models in practical applications. In
this paper, we address the problem of assessing and enhancing the robustness of
object detection models against natural perturbations, such as varying lighting
conditions, blur, and brightness. We analyze four state-of-the-art deep neural
network models, Detr-ResNet-101, Detr-ResNet-50, YOLOv4, and YOLOv4-tiny, using
the COCO 2017 dataset and ExDark dataset. By simulating synthetic perturbations
with the AugLy package, we systematically explore the optimal level of
synthetic perturbation required to improve the models robustness through data
augmentation techniques. Our comprehensive ablation study meticulously
evaluates the impact of synthetic perturbations on object detection models
performance against real-world distribution shifts, establishing a tangible
connection between synthetic augmentation and real-world robustness. Our
findings not only substantiate the effectiveness of synthetic perturbations in
improving model robustness, but also provide valuable insights for researchers
and practitioners in developing more robust and reliable object detection
models tailored for real-world applications.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 19:55:51 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Premakumara",
"Nilantha",
""
],
[
"Jalaian",
"Brian",
""
],
[
"Suri",
"Niranjan",
""
],
[
"Samani",
"Hooman",
""
]
] | TITLE: Enhancing object detection robustness: A synthetic and natural
perturbation approach
ABSTRACT: Robustness against real-world distribution shifts is crucial for the
successful deployment of object detection models in practical applications. In
this paper, we address the problem of assessing and enhancing the robustness of
object detection models against natural perturbations, such as varying lighting
conditions, blur, and brightness. We analyze four state-of-the-art deep neural
network models, Detr-ResNet-101, Detr-ResNet-50, YOLOv4, and YOLOv4-tiny, using
the COCO 2017 dataset and ExDark dataset. By simulating synthetic perturbations
with the AugLy package, we systematically explore the optimal level of
synthetic perturbation required to improve the models robustness through data
augmentation techniques. Our comprehensive ablation study meticulously
evaluates the impact of synthetic perturbations on object detection models
performance against real-world distribution shifts, establishing a tangible
connection between synthetic augmentation and real-world robustness. Our
findings not only substantiate the effectiveness of synthetic perturbations in
improving model robustness, but also provide valuable insights for researchers
and practitioners in developing more robust and reliable object detection
models tailored for real-world applications.
| no_new_dataset | 0.709636 |
2304.10628 | Hao Xiang | Hao Xiang, Runsheng Xu, Jiaqi Ma | HM-ViT: Hetero-modal Vehicle-to-Vehicle Cooperative perception with
vision transformer | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Vehicle-to-Vehicle technologies have enabled autonomous vehicles to share
information to see through occlusions, greatly enhancing perception
performance. Nevertheless, existing works all focused on homogeneous traffic
where vehicles are equipped with the same type of sensors, which significantly
hampers the scale of collaboration and benefit of cross-modality interactions.
In this paper, we investigate the multi-agent hetero-modal cooperative
perception problem where agents may have distinct sensor modalities. We present
HM-ViT, the first unified multi-agent hetero-modal cooperative perception
framework that can collaboratively predict 3D objects for highly dynamic
vehicle-to-vehicle (V2V) collaborations with varying numbers and types of
agents. To effectively fuse features from multi-view images and LiDAR point
clouds, we design a novel heterogeneous 3D graph transformer to jointly reason
inter-agent and intra-agent interactions. The extensive experiments on the V2V
perception dataset OPV2V demonstrate that the HM-ViT outperforms SOTA
cooperative perception methods for V2V hetero-modal cooperative perception. We
will release codes to facilitate future research.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 20:09:59 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Xiang",
"Hao",
""
],
[
"Xu",
"Runsheng",
""
],
[
"Ma",
"Jiaqi",
""
]
] | TITLE: HM-ViT: Hetero-modal Vehicle-to-Vehicle Cooperative perception with
vision transformer
ABSTRACT: Vehicle-to-Vehicle technologies have enabled autonomous vehicles to share
information to see through occlusions, greatly enhancing perception
performance. Nevertheless, existing works all focused on homogeneous traffic
where vehicles are equipped with the same type of sensors, which significantly
hampers the scale of collaboration and benefit of cross-modality interactions.
In this paper, we investigate the multi-agent hetero-modal cooperative
perception problem where agents may have distinct sensor modalities. We present
HM-ViT, the first unified multi-agent hetero-modal cooperative perception
framework that can collaboratively predict 3D objects for highly dynamic
vehicle-to-vehicle (V2V) collaborations with varying numbers and types of
agents. To effectively fuse features from multi-view images and LiDAR point
clouds, we design a novel heterogeneous 3D graph transformer to jointly reason
inter-agent and intra-agent interactions. The extensive experiments on the V2V
perception dataset OPV2V demonstrate that the HM-ViT outperforms SOTA
cooperative perception methods for V2V hetero-modal cooperative perception. We
will release codes to facilitate future research.
| new_dataset | 0.703728 |
2304.10642 | Anik Saha | Anik Saha, Alex Gittens, Bulent Yener | Word Sense Induction with Knowledge Distillation from BERT | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Pre-trained contextual language models are ubiquitously employed for language
understanding tasks, but are unsuitable for resource-constrained systems.
Noncontextual word embeddings are an efficient alternative in these settings.
Such methods typically use one vector to encode multiple different meanings of
a word, and incur errors due to polysemy. This paper proposes a two-stage
method to distill multiple word senses from a pre-trained language model (BERT)
by using attention over the senses of a word in a context and transferring this
sense information to fit multi-sense embeddings in a skip-gram-like framework.
We demonstrate an effective approach to training the sense disambiguation
mechanism in our model with a distribution over word senses extracted from the
output layer embeddings of BERT. Experiments on the contextual word similarity
and sense induction tasks show that this method is superior to or competitive
with state-of-the-art multi-sense embeddings on multiple benchmark data sets,
and experiments with an embedding-based topic model (ETM) demonstrates the
benefits of using this multi-sense embedding in a downstream application.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 21:05:35 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Saha",
"Anik",
""
],
[
"Gittens",
"Alex",
""
],
[
"Yener",
"Bulent",
""
]
] | TITLE: Word Sense Induction with Knowledge Distillation from BERT
ABSTRACT: Pre-trained contextual language models are ubiquitously employed for language
understanding tasks, but are unsuitable for resource-constrained systems.
Noncontextual word embeddings are an efficient alternative in these settings.
Such methods typically use one vector to encode multiple different meanings of
a word, and incur errors due to polysemy. This paper proposes a two-stage
method to distill multiple word senses from a pre-trained language model (BERT)
by using attention over the senses of a word in a context and transferring this
sense information to fit multi-sense embeddings in a skip-gram-like framework.
We demonstrate an effective approach to training the sense disambiguation
mechanism in our model with a distribution over word senses extracted from the
output layer embeddings of BERT. Experiments on the contextual word similarity
and sense induction tasks show that this method is superior to or competitive
with state-of-the-art multi-sense embeddings on multiple benchmark data sets,
and experiments with an embedding-based topic model (ETM) demonstrates the
benefits of using this multi-sense embedding in a downstream application.
| no_new_dataset | 0.710666 |
2304.10643 | Gezheng Wen | Chaitra Hedge, Gezheng Wen, Layne C. Price | Activity Classification Using Unsupervised Domain Transfer from Body
Worn Sensors | null | null | null | null | cs.LG cs.HC eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Activity classification has become a vital feature of wearable health
tracking devices. As innovation in this field grows, wearable devices worn on
different parts of the body are emerging. To perform activity classification on
a new body location, labeled data corresponding to the new locations are
generally required, but this is expensive to acquire. In this work, we present
an innovative method to leverage an existing activity classifier, trained on
Inertial Measurement Unit (IMU) data from a reference body location (the source
domain), in order to perform activity classification on a new body location
(the target domain) in an unsupervised way, i.e. without the need for
classification labels at the new location. Specifically, given an IMU embedding
model trained to perform activity classification at the source domain, we train
an embedding model to perform activity classification at the target domain by
replicating the embeddings at the source domain. This is achieved using
simultaneous IMU measurements at the source and target domains. The replicated
embeddings at the target domain are used by a classification model that has
previously been trained on the source domain to perform activity classification
at the target domain. We have evaluated the proposed methods on three activity
classification datasets PAMAP2, MHealth, and Opportunity, yielding high F1
scores of 67.19%, 70.40% and 68.34%, respectively when the source domain is the
wrist and the target domain is the torso.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 21:08:00 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Hedge",
"Chaitra",
""
],
[
"Wen",
"Gezheng",
""
],
[
"Price",
"Layne C.",
""
]
] | TITLE: Activity Classification Using Unsupervised Domain Transfer from Body
Worn Sensors
ABSTRACT: Activity classification has become a vital feature of wearable health
tracking devices. As innovation in this field grows, wearable devices worn on
different parts of the body are emerging. To perform activity classification on
a new body location, labeled data corresponding to the new locations are
generally required, but this is expensive to acquire. In this work, we present
an innovative method to leverage an existing activity classifier, trained on
Inertial Measurement Unit (IMU) data from a reference body location (the source
domain), in order to perform activity classification on a new body location
(the target domain) in an unsupervised way, i.e. without the need for
classification labels at the new location. Specifically, given an IMU embedding
model trained to perform activity classification at the source domain, we train
an embedding model to perform activity classification at the target domain by
replicating the embeddings at the source domain. This is achieved using
simultaneous IMU measurements at the source and target domains. The replicated
embeddings at the target domain are used by a classification model that has
previously been trained on the source domain to perform activity classification
at the target domain. We have evaluated the proposed methods on three activity
classification datasets PAMAP2, MHealth, and Opportunity, yielding high F1
scores of 67.19%, 70.40% and 68.34%, respectively when the source domain is the
wrist and the target domain is the torso.
| no_new_dataset | 0.713032 |
2304.10655 | Anna Meyer | Anna P. Meyer, Aws Albarghouthi, Loris D'Antoni | The Dataset Multiplicity Problem: How Unreliable Data Impacts
Predictions | 25 pages, 8 figures. Accepted at FAccT '23 | null | null | null | cs.LG cs.CY | http://creativecommons.org/licenses/by/4.0/ | We introduce dataset multiplicity, a way to study how inaccuracies,
uncertainty, and social bias in training datasets impact test-time predictions.
The dataset multiplicity framework asks a counterfactual question of what the
set of resultant models (and associated test-time predictions) would be if we
could somehow access all hypothetical, unbiased versions of the dataset. We
discuss how to use this framework to encapsulate various sources of uncertainty
in datasets' factualness, including systemic social bias, data collection
practices, and noisy labels or features. We show how to exactly analyze the
impacts of dataset multiplicity for a specific model architecture and type of
uncertainty: linear models with label errors. Our empirical analysis shows that
real-world datasets, under reasonable assumptions, contain many test samples
whose predictions are affected by dataset multiplicity. Furthermore, the choice
of domain-specific dataset multiplicity definition determines what samples are
affected, and whether different demographic groups are disparately impacted.
Finally, we discuss implications of dataset multiplicity for machine learning
practice and research, including considerations for when model outcomes should
not be trusted.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 21:31:15 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Meyer",
"Anna P.",
""
],
[
"Albarghouthi",
"Aws",
""
],
[
"D'Antoni",
"Loris",
""
]
] | TITLE: The Dataset Multiplicity Problem: How Unreliable Data Impacts
Predictions
ABSTRACT: We introduce dataset multiplicity, a way to study how inaccuracies,
uncertainty, and social bias in training datasets impact test-time predictions.
The dataset multiplicity framework asks a counterfactual question of what the
set of resultant models (and associated test-time predictions) would be if we
could somehow access all hypothetical, unbiased versions of the dataset. We
discuss how to use this framework to encapsulate various sources of uncertainty
in datasets' factualness, including systemic social bias, data collection
practices, and noisy labels or features. We show how to exactly analyze the
impacts of dataset multiplicity for a specific model architecture and type of
uncertainty: linear models with label errors. Our empirical analysis shows that
real-world datasets, under reasonable assumptions, contain many test samples
whose predictions are affected by dataset multiplicity. Furthermore, the choice
of domain-specific dataset multiplicity definition determines what samples are
affected, and whether different demographic groups are disparately impacted.
Finally, we discuss implications of dataset multiplicity for machine learning
practice and research, including considerations for when model outcomes should
not be trusted.
| no_new_dataset | 0.708231 |
2304.10668 | Costas Mavromatis | Costas Mavromatis, Vassilis N. Ioannidis, Shen Wang, Da Zheng, Soji
Adeshina, Jun Ma, Han Zhao, Christos Faloutsos, George Karypis | Train Your Own GNN Teacher: Graph-Aware Distillation on Textual Graphs | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | How can we learn effective node representations on textual graphs? Graph
Neural Networks (GNNs) that use Language Models (LMs) to encode textual
information of graphs achieve state-of-the-art performance in many node
classification tasks. Yet, combining GNNs with LMs has not been widely explored
for practical deployments due to its scalability issues. In this work, we
tackle this challenge by developing a Graph-Aware Distillation framework (GRAD)
to encode graph structures into an LM for graph-free, fast inference. Different
from conventional knowledge distillation, GRAD jointly optimizes a GNN teacher
and a graph-free student over the graph's nodes via a shared LM. This
encourages the graph-free student to exploit graph information encoded by the
GNN teacher while at the same time, enables the GNN teacher to better leverage
textual information from unlabeled nodes. As a result, the teacher and the
student models learn from each other to improve their overall performance.
Experiments in eight node classification benchmarks in both transductive and
inductive settings showcase GRAD's superiority over existing distillation
approaches for textual graphs.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 22:34:20 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Mavromatis",
"Costas",
""
],
[
"Ioannidis",
"Vassilis N.",
""
],
[
"Wang",
"Shen",
""
],
[
"Zheng",
"Da",
""
],
[
"Adeshina",
"Soji",
""
],
[
"Ma",
"Jun",
""
],
[
"Zhao",
"Han",
""
],
[
"Faloutsos",
"Christos",
""
],
[
"Karypis",
"George",
""
]
] | TITLE: Train Your Own GNN Teacher: Graph-Aware Distillation on Textual Graphs
ABSTRACT: How can we learn effective node representations on textual graphs? Graph
Neural Networks (GNNs) that use Language Models (LMs) to encode textual
information of graphs achieve state-of-the-art performance in many node
classification tasks. Yet, combining GNNs with LMs has not been widely explored
for practical deployments due to its scalability issues. In this work, we
tackle this challenge by developing a Graph-Aware Distillation framework (GRAD)
to encode graph structures into an LM for graph-free, fast inference. Different
from conventional knowledge distillation, GRAD jointly optimizes a GNN teacher
and a graph-free student over the graph's nodes via a shared LM. This
encourages the graph-free student to exploit graph information encoded by the
GNN teacher while at the same time, enables the GNN teacher to better leverage
textual information from unlabeled nodes. As a result, the teacher and the
student models learn from each other to improve their overall performance.
Experiments in eight node classification benchmarks in both transductive and
inductive settings showcase GRAD's superiority over existing distillation
approaches for textual graphs.
| no_new_dataset | 0.710057 |
2304.10677 | Abdurrahman Gumus | Ceyhun Efe Kayan, Talha Enes Koksal, Arda Sevinc, Abdurrahman Gumus | Deep reproductive feature generation framework for the diagnosis of
COVID-19 and viral pneumonia using chest X-ray images | null | null | null | null | eess.IV cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid and accurate detection of COVID-19 cases is critical for timely
treatment and preventing the spread of the disease. In this study, a two-stage
feature extraction framework using eight state-of-the-art pre-trained deep
Convolutional Neural Networks (CNNs) and an autoencoder is proposed to
determine the health conditions of patients (COVID-19, Normal, Viral Pneumonia)
based on chest X-rays. The X-ray scans are divided into four equally sized
sections and analyzed by deep pre-trained CNNs. Subsequently, an autoencoder
with three hidden layers is trained to extract reproductive features from the
concatenated ouput of CNNs. To evaluate the performance of the proposed
framework, three different classifiers, which are single-layer perceptron
(SLP), multi-layer perceptron (MLP), and support vector machine (SVM) are used.
Furthermore, the deep CNN architectures are used to create benchmark models and
trained on the same dataset for comparision. The proposed framework outperforms
other frameworks wih pre-trained feature extractors in binary classification
and shows competitive results in three-class classification. The proposed
methodology is task-independent and suitable for addressing various problems.
The results show that the discriminative features are a subset of the
reproductive features, suggesting that extracting task-independent features is
superior to the extraction only task-based features. The flexibility and
task-independence of the reproductive features make the conceptive information
approach more favorable. The proposed methodology is novel and shows promising
results for analyzing medical image data.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 23:52:21 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Kayan",
"Ceyhun Efe",
""
],
[
"Koksal",
"Talha Enes",
""
],
[
"Sevinc",
"Arda",
""
],
[
"Gumus",
"Abdurrahman",
""
]
] | TITLE: Deep reproductive feature generation framework for the diagnosis of
COVID-19 and viral pneumonia using chest X-ray images
ABSTRACT: The rapid and accurate detection of COVID-19 cases is critical for timely
treatment and preventing the spread of the disease. In this study, a two-stage
feature extraction framework using eight state-of-the-art pre-trained deep
Convolutional Neural Networks (CNNs) and an autoencoder is proposed to
determine the health conditions of patients (COVID-19, Normal, Viral Pneumonia)
based on chest X-rays. The X-ray scans are divided into four equally sized
sections and analyzed by deep pre-trained CNNs. Subsequently, an autoencoder
with three hidden layers is trained to extract reproductive features from the
concatenated ouput of CNNs. To evaluate the performance of the proposed
framework, three different classifiers, which are single-layer perceptron
(SLP), multi-layer perceptron (MLP), and support vector machine (SVM) are used.
Furthermore, the deep CNN architectures are used to create benchmark models and
trained on the same dataset for comparision. The proposed framework outperforms
other frameworks wih pre-trained feature extractors in binary classification
and shows competitive results in three-class classification. The proposed
methodology is task-independent and suitable for addressing various problems.
The results show that the discriminative features are a subset of the
reproductive features, suggesting that extracting task-independent features is
superior to the extraction only task-based features. The flexibility and
task-independence of the reproductive features make the conceptive information
approach more favorable. The proposed methodology is novel and shows promising
results for analyzing medical image data.
| no_new_dataset | 0.712595 |
2304.10687 | Huiyu Gao | Huiyu Gao, Wei Mao, Miaomiao Liu | VisFusion: Visibility-aware Online 3D Scene Reconstruction from Videos | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose VisFusion, a visibility-aware online 3D scene reconstruction
approach from posed monocular videos. In particular, we aim to reconstruct the
scene from volumetric features. Unlike previous reconstruction methods which
aggregate features for each voxel from input views without considering its
visibility, we aim to improve the feature fusion by explicitly inferring its
visibility from a similarity matrix, computed from its projected features in
each image pair. Following previous works, our model is a coarse-to-fine
pipeline including a volume sparsification process. Different from their works
which sparsify voxels globally with a fixed occupancy threshold, we perform the
sparsification on a local feature volume along each visual ray to preserve at
least one voxel per ray for more fine details. The sparse local volume is then
fused with a global one for online reconstruction. We further propose to
predict TSDF in a coarse-to-fine manner by learning its residuals across scales
leading to better TSDF predictions. Experimental results on benchmarks show
that our method can achieve superior performance with more scene details. Code
is available at: https://github.com/huiyu-gao/VisFusion
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 00:47:05 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Gao",
"Huiyu",
""
],
[
"Mao",
"Wei",
""
],
[
"Liu",
"Miaomiao",
""
]
] | TITLE: VisFusion: Visibility-aware Online 3D Scene Reconstruction from Videos
ABSTRACT: We propose VisFusion, a visibility-aware online 3D scene reconstruction
approach from posed monocular videos. In particular, we aim to reconstruct the
scene from volumetric features. Unlike previous reconstruction methods which
aggregate features for each voxel from input views without considering its
visibility, we aim to improve the feature fusion by explicitly inferring its
visibility from a similarity matrix, computed from its projected features in
each image pair. Following previous works, our model is a coarse-to-fine
pipeline including a volume sparsification process. Different from their works
which sparsify voxels globally with a fixed occupancy threshold, we perform the
sparsification on a local feature volume along each visual ray to preserve at
least one voxel per ray for more fine details. The sparse local volume is then
fused with a global one for online reconstruction. We further propose to
predict TSDF in a coarse-to-fine manner by learning its residuals across scales
leading to better TSDF predictions. Experimental results on benchmarks show
that our method can achieve superior performance with more scene details. Code
is available at: https://github.com/huiyu-gao/VisFusion
| no_new_dataset | 0.711469 |
2304.10693 | Ping Jiang | Ping Jiang, Junji Oaki, Yoshiyuki Ishihara, Junichiro Ooga | Multiple-object Grasping Using a Multiple-suction-cup Vacuum Gripper in
Cluttered Scenes | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multiple-suction-cup grasping can improve the efficiency of bin picking in
cluttered scenes. In this paper, we propose a grasp planner for a vacuum
gripper to use multiple suction cups to simultaneously grasp multiple objects
or an object with a large surface. To take on the challenge of determining
where to grasp and which cups to activate when grasping, we used 3D convolution
to convolve the affordable areas inferred by neural network with the gripper
kernel in order to find graspable positions of sampled gripper orientations.
The kernel used for 3D convolution in this work was encoded including cup ID
information, which helps to directly determine which cups to activate by
decoding the convolution results. Furthermore, a sorting algorithm is proposed
to find the optimal grasp among the candidates. Our planner exhibited good
generality and successfully found multiple-cup grasps in previous affordance
map datasets. Our planner also exhibited improved picking efficiency using
multiple suction cups in physical robot picking experiments. Compared with
single-object (single-cup) grasping, multiple-cup grasping contributed to
1.45x, 1.65x, and 1.16x increases in efficiency for picking boxes, fruits, and
daily necessities, respectively.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 01:26:54 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Jiang",
"Ping",
""
],
[
"Oaki",
"Junji",
""
],
[
"Ishihara",
"Yoshiyuki",
""
],
[
"Ooga",
"Junichiro",
""
]
] | TITLE: Multiple-object Grasping Using a Multiple-suction-cup Vacuum Gripper in
Cluttered Scenes
ABSTRACT: Multiple-suction-cup grasping can improve the efficiency of bin picking in
cluttered scenes. In this paper, we propose a grasp planner for a vacuum
gripper to use multiple suction cups to simultaneously grasp multiple objects
or an object with a large surface. To take on the challenge of determining
where to grasp and which cups to activate when grasping, we used 3D convolution
to convolve the affordable areas inferred by neural network with the gripper
kernel in order to find graspable positions of sampled gripper orientations.
The kernel used for 3D convolution in this work was encoded including cup ID
information, which helps to directly determine which cups to activate by
decoding the convolution results. Furthermore, a sorting algorithm is proposed
to find the optimal grasp among the candidates. Our planner exhibited good
generality and successfully found multiple-cup grasps in previous affordance
map datasets. Our planner also exhibited improved picking efficiency using
multiple suction cups in physical robot picking experiments. Compared with
single-object (single-cup) grasping, multiple-cup grasping contributed to
1.45x, 1.65x, and 1.16x increases in efficiency for picking boxes, fruits, and
daily necessities, respectively.
| no_new_dataset | 0.713276 |
2304.10704 | Guanchu Wang | Guanchu Wang and Ninghao Liu and Daochen Zha and Xia Hu | Interactive System-wise Anomaly Detection | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Anomaly detection, where data instances are discovered containing feature
patterns different from the majority, plays a fundamental role in various
applications. However, it is challenging for existing methods to handle the
scenarios where the instances are systems whose characteristics are not readily
observed as data. Appropriate interactions are needed to interact with the
systems and identify those with abnormal responses. Detecting system-wise
anomalies is a challenging task due to several reasons including: how to
formally define the system-wise anomaly detection problem; how to find the
effective activation signal for interacting with systems to progressively
collect the data and learn the detector; how to guarantee stable training in
such a non-stationary scenario with real-time interactions? To address the
challenges, we propose InterSAD (Interactive System-wise Anomaly Detection).
Specifically, first, we adopt Markov decision process to model the interactive
systems, and define anomalous systems as anomalous transition and anomalous
reward systems. Then, we develop an end-to-end approach which includes an
encoder-decoder module that learns system embeddings, and a policy network to
generate effective activation for separating embeddings of normal and anomaly
systems. Finally, we design a training method to stabilize the learning
process, which includes a replay buffer to store historical interaction data
and allow them to be re-sampled. Experiments on two benchmark environments,
including identifying the anomalous robotic systems and detecting user data
poisoning in recommendation models, demonstrate the superiority of InterSAD
compared with state-of-the-art baselines methods.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 02:20:24 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Wang",
"Guanchu",
""
],
[
"Liu",
"Ninghao",
""
],
[
"Zha",
"Daochen",
""
],
[
"Hu",
"Xia",
""
]
] | TITLE: Interactive System-wise Anomaly Detection
ABSTRACT: Anomaly detection, where data instances are discovered containing feature
patterns different from the majority, plays a fundamental role in various
applications. However, it is challenging for existing methods to handle the
scenarios where the instances are systems whose characteristics are not readily
observed as data. Appropriate interactions are needed to interact with the
systems and identify those with abnormal responses. Detecting system-wise
anomalies is a challenging task due to several reasons including: how to
formally define the system-wise anomaly detection problem; how to find the
effective activation signal for interacting with systems to progressively
collect the data and learn the detector; how to guarantee stable training in
such a non-stationary scenario with real-time interactions? To address the
challenges, we propose InterSAD (Interactive System-wise Anomaly Detection).
Specifically, first, we adopt Markov decision process to model the interactive
systems, and define anomalous systems as anomalous transition and anomalous
reward systems. Then, we develop an end-to-end approach which includes an
encoder-decoder module that learns system embeddings, and a policy network to
generate effective activation for separating embeddings of normal and anomaly
systems. Finally, we design a training method to stabilize the learning
process, which includes a replay buffer to store historical interaction data
and allow them to be re-sampled. Experiments on two benchmark environments,
including identifying the anomalous robotic systems and detecting user data
poisoning in recommendation models, demonstrate the superiority of InterSAD
compared with state-of-the-art baselines methods.
| no_new_dataset | 0.70878 |
2304.10705 | Houcheng Su | Houcheng Su, Jintao Huang, Daixian Liu, Rui Yan, Jiao Li, Chi-man Vong | Graph based Label Enhancement for Multi-instance Multi-label learning | 7 pages,2 figures | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Multi-instance multi-label (MIML) learning is widely applicated in numerous
domains, such as the image classification where one image contains multiple
instances correlated with multiple logic labels simultaneously. The related
labels in existing MIML are all assumed as logical labels with equal
significance. However, in practical applications in MIML, significance of each
label for multiple instances per bag (such as an image) is significant
different. Ignoring labeling significance will greatly lose the semantic
information of the object, so that MIML is not applicable in complex scenes
with a poor learning performance. To this end, this paper proposed a novel MIML
framework based on graph label enhancement, namely GLEMIML, to improve the
classification performance of MIML by leveraging label significance. GLEMIML
first recognizes the correlations among instances by establishing the graph and
then migrates the implicit information mined from the feature space to the
label space via nonlinear mapping, thus recovering the label significance.
Finally, GLEMIML is trained on the enhanced data through matching and
interaction mechanisms. GLEMIML (AvgRank: 1.44) can effectively improve the
performance of MIML by mining the label distribution mechanism and show better
results than the SOTA method (AvgRank: 2.92) on multiple benchmark datasets.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 02:24:49 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Su",
"Houcheng",
""
],
[
"Huang",
"Jintao",
""
],
[
"Liu",
"Daixian",
""
],
[
"Yan",
"Rui",
""
],
[
"Li",
"Jiao",
""
],
[
"Vong",
"Chi-man",
""
]
] | TITLE: Graph based Label Enhancement for Multi-instance Multi-label learning
ABSTRACT: Multi-instance multi-label (MIML) learning is widely applicated in numerous
domains, such as the image classification where one image contains multiple
instances correlated with multiple logic labels simultaneously. The related
labels in existing MIML are all assumed as logical labels with equal
significance. However, in practical applications in MIML, significance of each
label for multiple instances per bag (such as an image) is significant
different. Ignoring labeling significance will greatly lose the semantic
information of the object, so that MIML is not applicable in complex scenes
with a poor learning performance. To this end, this paper proposed a novel MIML
framework based on graph label enhancement, namely GLEMIML, to improve the
classification performance of MIML by leveraging label significance. GLEMIML
first recognizes the correlations among instances by establishing the graph and
then migrates the implicit information mined from the feature space to the
label space via nonlinear mapping, thus recovering the label significance.
Finally, GLEMIML is trained on the enhanced data through matching and
interaction mechanisms. GLEMIML (AvgRank: 1.44) can effectively improve the
performance of MIML by mining the label distribution mechanism and show better
results than the SOTA method (AvgRank: 2.92) on multiple benchmark datasets.
| no_new_dataset | 0.708994 |
2304.10706 | Xiaosong Yuan | Xiaosong Yuan, Ke Chen, Wanli Zuo, Yijia Zhang | TC-GAT: Graph Attention Network for Temporal Causality Discovery | Accepted by IJCNN 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The present study explores the intricacies of causal relationship extraction,
a vital component in the pursuit of causality knowledge. Causality is
frequently intertwined with temporal elements, as the progression from cause to
effect is not instantaneous but rather ensconced in a temporal dimension. Thus,
the extraction of temporal causality holds paramount significance in the field.
In light of this, we propose a method for extracting causality from the text
that integrates both temporal and causal relations, with a particular focus on
the time aspect. To this end, we first compile a dataset that encompasses
temporal relationships. Subsequently, we present a novel model, TC-GAT, which
employs a graph attention mechanism to assign weights to the temporal
relationships and leverages a causal knowledge graph to determine the adjacency
matrix. Additionally, we implement an equilibrium mechanism to regulate the
interplay between temporal and causal relations. Our experiments demonstrate
that our proposed method significantly surpasses baseline models in the task of
causality extraction.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 02:26:42 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Yuan",
"Xiaosong",
""
],
[
"Chen",
"Ke",
""
],
[
"Zuo",
"Wanli",
""
],
[
"Zhang",
"Yijia",
""
]
] | TITLE: TC-GAT: Graph Attention Network for Temporal Causality Discovery
ABSTRACT: The present study explores the intricacies of causal relationship extraction,
a vital component in the pursuit of causality knowledge. Causality is
frequently intertwined with temporal elements, as the progression from cause to
effect is not instantaneous but rather ensconced in a temporal dimension. Thus,
the extraction of temporal causality holds paramount significance in the field.
In light of this, we propose a method for extracting causality from the text
that integrates both temporal and causal relations, with a particular focus on
the time aspect. To this end, we first compile a dataset that encompasses
temporal relationships. Subsequently, we present a novel model, TC-GAT, which
employs a graph attention mechanism to assign weights to the temporal
relationships and leverages a causal knowledge graph to determine the adjacency
matrix. Additionally, we implement an equilibrium mechanism to regulate the
interplay between temporal and causal relations. Our experiments demonstrate
that our proposed method significantly surpasses baseline models in the task of
causality extraction.
| new_dataset | 0.716411 |
2304.10719 | Yuxuan Liu | Yuxuan Liu, Zhenhua Xu, Huaiyang Huang, Lujia Wang, Ming Liu | FSNet: Redesign Self-Supervised MonoDepth for Full-Scale Depth
Prediction for Autonomous Driving | 12 pages. conditionally accepted by IEEE T-ASE | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Predicting accurate depth with monocular images is important for low-cost
robotic applications and autonomous driving. This study proposes a
comprehensive self-supervised framework for accurate scale-aware depth
prediction on autonomous driving scenes utilizing inter-frame poses obtained
from inertial measurements. In particular, we introduce a Full-Scale depth
prediction network named FSNet. FSNet contains four important improvements over
existing self-supervised models: (1) a multichannel output representation for
stable training of depth prediction in driving scenarios, (2) an
optical-flow-based mask designed for dynamic object removal, (3) a
self-distillation training strategy to augment the training process, and (4) an
optimization-based post-processing algorithm in test time, fusing the results
from visual odometry. With this framework, robots and vehicles with only one
well-calibrated camera can collect sequences of training image frames and
camera poses, and infer accurate 3D depths of the environment without extra
labeling work or 3D data. Extensive experiments on the KITTI dataset, KITTI-360
dataset and the nuScenes dataset demonstrate the potential of FSNet. More
visualizations are presented in \url{https://sites.google.com/view/fsnet/home}
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 03:17:04 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Liu",
"Yuxuan",
""
],
[
"Xu",
"Zhenhua",
""
],
[
"Huang",
"Huaiyang",
""
],
[
"Wang",
"Lujia",
""
],
[
"Liu",
"Ming",
""
]
] | TITLE: FSNet: Redesign Self-Supervised MonoDepth for Full-Scale Depth
Prediction for Autonomous Driving
ABSTRACT: Predicting accurate depth with monocular images is important for low-cost
robotic applications and autonomous driving. This study proposes a
comprehensive self-supervised framework for accurate scale-aware depth
prediction on autonomous driving scenes utilizing inter-frame poses obtained
from inertial measurements. In particular, we introduce a Full-Scale depth
prediction network named FSNet. FSNet contains four important improvements over
existing self-supervised models: (1) a multichannel output representation for
stable training of depth prediction in driving scenarios, (2) an
optical-flow-based mask designed for dynamic object removal, (3) a
self-distillation training strategy to augment the training process, and (4) an
optimization-based post-processing algorithm in test time, fusing the results
from visual odometry. With this framework, robots and vehicles with only one
well-calibrated camera can collect sequences of training image frames and
camera poses, and infer accurate 3D depths of the environment without extra
labeling work or 3D data. Extensive experiments on the KITTI dataset, KITTI-360
dataset and the nuScenes dataset demonstrate the potential of FSNet. More
visualizations are presented in \url{https://sites.google.com/view/fsnet/home}
| no_new_dataset | 0.710465 |
2304.10724 | Koki Ikeda | Koki Ikeda and Isao Ono | Natural Evolution Strategy for Mixed-Integer Black-Box Optimization | Accepted for GECCO'23 | null | 10.1145/3583131.3590518 | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper proposes a natural evolution strategy (NES) for mixed-integer
black-box optimization (MI-BBO) that appears in real-world problems such as
hyperparameter optimization of machine learning and materials design. This
problem is difficult to optimize because plateaus where the values do not
change appear when the integer variables are relaxed to the continuous ones.
CMA-ES w. Margin that addresses the plateaus reportedly showed good performance
on MI-BBO benchmark problems. However, it has been observed that the search
performance of CMA-ES w. Margin deteriorates when continuous variables
contribute more to the objective function value than integer ones. In order to
address the problem of CMA-ES w. Margin, we propose Distance-weighted
eXponential Natural Evolution Strategy taking account of Implicit Constraint
and Integer (DX-NES-ICI). We compare the search performance of DX-NES-ICI with
that of CMA-ES w. Margin through numerical experiments. As a result, DX-NES-ICI
was up to 3.7 times better than CMA-ES w. Margin in terms of a rate of finding
the optimal solutions on benchmark problems where continuous variables
contribute more to the objective function value than integer ones. DX-NES-ICI
also outperformed CMA-ES w. Margin on problems where CMA-ES w. Margin
originally showed good performance.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 03:34:22 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Ikeda",
"Koki",
""
],
[
"Ono",
"Isao",
""
]
] | TITLE: Natural Evolution Strategy for Mixed-Integer Black-Box Optimization
ABSTRACT: This paper proposes a natural evolution strategy (NES) for mixed-integer
black-box optimization (MI-BBO) that appears in real-world problems such as
hyperparameter optimization of machine learning and materials design. This
problem is difficult to optimize because plateaus where the values do not
change appear when the integer variables are relaxed to the continuous ones.
CMA-ES w. Margin that addresses the plateaus reportedly showed good performance
on MI-BBO benchmark problems. However, it has been observed that the search
performance of CMA-ES w. Margin deteriorates when continuous variables
contribute more to the objective function value than integer ones. In order to
address the problem of CMA-ES w. Margin, we propose Distance-weighted
eXponential Natural Evolution Strategy taking account of Implicit Constraint
and Integer (DX-NES-ICI). We compare the search performance of DX-NES-ICI with
that of CMA-ES w. Margin through numerical experiments. As a result, DX-NES-ICI
was up to 3.7 times better than CMA-ES w. Margin in terms of a rate of finding
the optimal solutions on benchmark problems where continuous variables
contribute more to the objective function value than integer ones. DX-NES-ICI
also outperformed CMA-ES w. Margin on problems where CMA-ES w. Margin
originally showed good performance.
| no_new_dataset | 0.711606 |
2304.10733 | Zhiwei Wei | Wei Zhiwei, Xiao Yi, Tong Ying, Xu Wenjia, Wang Yang | Linear building pattern recognition via spatial knowledge graph | in Chinese language | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Building patterns are important urban structures that reflect the effect of
the urban material and social-economic on a region. Previous researches are
mostly based on the graph isomorphism method and use rules to recognize
building patterns, which are not efficient. The knowledge graph uses the graph
to model the relationship between entities, and specific subgraph patterns can
be efficiently obtained by using relevant reasoning tools. Thus, we try to
apply the knowledge graph to recognize linear building patterns. First, we use
the property graph to express the spatial relations in proximity, similar and
linear arrangement between buildings; secondly, the rules of linear pattern
recognition are expressed as the rules of knowledge graph reasoning; finally,
the linear building patterns are recognized by using the rule-based reasoning
in the built knowledge graph. The experimental results on a dataset containing
1289 buildings show that the method in this paper can achieve the same
precision and recall as the existing methods; meanwhile, the recognition
efficiency is improved by 5.98 times.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 04:05:02 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Zhiwei",
"Wei",
""
],
[
"Yi",
"Xiao",
""
],
[
"Ying",
"Tong",
""
],
[
"Wenjia",
"Xu",
""
],
[
"Yang",
"Wang",
""
]
] | TITLE: Linear building pattern recognition via spatial knowledge graph
ABSTRACT: Building patterns are important urban structures that reflect the effect of
the urban material and social-economic on a region. Previous researches are
mostly based on the graph isomorphism method and use rules to recognize
building patterns, which are not efficient. The knowledge graph uses the graph
to model the relationship between entities, and specific subgraph patterns can
be efficiently obtained by using relevant reasoning tools. Thus, we try to
apply the knowledge graph to recognize linear building patterns. First, we use
the property graph to express the spatial relations in proximity, similar and
linear arrangement between buildings; secondly, the rules of linear pattern
recognition are expressed as the rules of knowledge graph reasoning; finally,
the linear building patterns are recognized by using the rule-based reasoning
in the built knowledge graph. The experimental results on a dataset containing
1289 buildings show that the method in this paper can achieve the same
precision and recall as the existing methods; meanwhile, the recognition
efficiency is improved by 5.98 times.
| no_new_dataset | 0.709629 |
2304.10736 | Zitong Lu | Zitong Lu and Julie D. Golomb | Generate your neural signals from mine: individual-to-individual EEG
converters | Proceedings of the 45th Annual Meeting of the Cognitive Science
Society (CogSci 2023) | null | null | null | q-bio.NC cs.CV cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Most models in cognitive and computational neuroscience trained on one
subject do not generalize to other subjects due to individual differences. An
ideal individual-to-individual neural converter is expected to generate real
neural signals of one subject from those of another one, which can overcome the
problem of individual differences for cognitive and computational models. In
this study, we propose a novel individual-to-individual EEG converter, called
EEG2EEG, inspired by generative models in computer vision. We applied THINGS
EEG2 dataset to train and test 72 independent EEG2EEG models corresponding to
72 pairs across 9 subjects. Our results demonstrate that EEG2EEG is able to
effectively learn the mapping of neural representations in EEG signals from one
subject to another and achieve high conversion performance. Additionally, the
generated EEG signals contain clearer representations of visual information
than that can be obtained from real data. This method establishes a novel and
state-of-the-art framework for neural conversion of EEG signals, which can
realize a flexible and high-performance mapping from individual to individual
and provide insight for both neural engineering and cognitive neuroscience.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 04:13:16 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Lu",
"Zitong",
""
],
[
"Golomb",
"Julie D.",
""
]
] | TITLE: Generate your neural signals from mine: individual-to-individual EEG
converters
ABSTRACT: Most models in cognitive and computational neuroscience trained on one
subject do not generalize to other subjects due to individual differences. An
ideal individual-to-individual neural converter is expected to generate real
neural signals of one subject from those of another one, which can overcome the
problem of individual differences for cognitive and computational models. In
this study, we propose a novel individual-to-individual EEG converter, called
EEG2EEG, inspired by generative models in computer vision. We applied THINGS
EEG2 dataset to train and test 72 independent EEG2EEG models corresponding to
72 pairs across 9 subjects. Our results demonstrate that EEG2EEG is able to
effectively learn the mapping of neural representations in EEG signals from one
subject to another and achieve high conversion performance. Additionally, the
generated EEG signals contain clearer representations of visual information
than that can be obtained from real data. This method establishes a novel and
state-of-the-art framework for neural conversion of EEG signals, which can
realize a flexible and high-performance mapping from individual to individual
and provide insight for both neural engineering and cognitive neuroscience.
| no_new_dataset | 0.70978 |
2304.10739 | Donghee Choi | Donghee Choi, Mogan Gim, Samy Badreddine, Hajung Kim, Donghyeon Park,
Jaewoo Kang | KitchenScale: Learning to predict ingredient quantities from recipe
contexts | Expert Systems with Applications 2023, Demo:
http://kitchenscale.korea.ac.kr/ | Expert Systems with Applications, Volume 224, 15 August 2023,
120041 | 10.1016/j.eswa.2023.120041 | null | cs.CL cs.AI cs.LG cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | Determining proper quantities for ingredients is an essential part of cooking
practice from the perspective of enriching tastiness and promoting healthiness.
We introduce KitchenScale, a fine-tuned Pre-trained Language Model (PLM) that
predicts a target ingredient's quantity and measurement unit given its recipe
context. To effectively train our KitchenScale model, we formulate an
ingredient quantity prediction task that consists of three sub-tasks which are
ingredient measurement type classification, unit classification, and quantity
regression task. Furthermore, we utilized transfer learning of cooking
knowledge from recipe texts to PLMs. We adopted the Discrete Latent Exponent
(DExp) method to cope with high variance of numerical scales in recipe corpora.
Experiments with our newly constructed dataset and recommendation examples
demonstrate KitchenScale's understanding of various recipe contexts and
generalizability in predicting ingredient quantities. We implemented a web
application for KitchenScale to demonstrate its functionality in recommending
ingredient quantities expressed in numerals (e.g., 2) with units (e.g., ounce).
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 04:28:16 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Choi",
"Donghee",
""
],
[
"Gim",
"Mogan",
""
],
[
"Badreddine",
"Samy",
""
],
[
"Kim",
"Hajung",
""
],
[
"Park",
"Donghyeon",
""
],
[
"Kang",
"Jaewoo",
""
]
] | TITLE: KitchenScale: Learning to predict ingredient quantities from recipe
contexts
ABSTRACT: Determining proper quantities for ingredients is an essential part of cooking
practice from the perspective of enriching tastiness and promoting healthiness.
We introduce KitchenScale, a fine-tuned Pre-trained Language Model (PLM) that
predicts a target ingredient's quantity and measurement unit given its recipe
context. To effectively train our KitchenScale model, we formulate an
ingredient quantity prediction task that consists of three sub-tasks which are
ingredient measurement type classification, unit classification, and quantity
regression task. Furthermore, we utilized transfer learning of cooking
knowledge from recipe texts to PLMs. We adopted the Discrete Latent Exponent
(DExp) method to cope with high variance of numerical scales in recipe corpora.
Experiments with our newly constructed dataset and recommendation examples
demonstrate KitchenScale's understanding of various recipe contexts and
generalizability in predicting ingredient quantities. We implemented a web
application for KitchenScale to demonstrate its functionality in recommending
ingredient quantities expressed in numerals (e.g., 2) with units (e.g., ounce).
| new_dataset | 0.712807 |
2304.10756 | Harsh Maheshwari | Harsh Maheshwari, Yen-Cheng Liu, Zsolt Kira | Missing Modality Robustness in Semi-Supervised Multi-Modal Semantic
Segmentation | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Using multiple spatial modalities has been proven helpful in improving
semantic segmentation performance. However, there are several real-world
challenges that have yet to be addressed: (a) improving label efficiency and
(b) enhancing robustness in realistic scenarios where modalities are missing at
the test time. To address these challenges, we first propose a simple yet
efficient multi-modal fusion mechanism Linear Fusion, that performs better than
the state-of-the-art multi-modal models even with limited supervision. Second,
we propose M3L: Multi-modal Teacher for Masked Modality Learning, a
semi-supervised framework that not only improves the multi-modal performance
but also makes the model robust to the realistic missing modality scenario
using unlabeled data. We create the first benchmark for semi-supervised
multi-modal semantic segmentation and also report the robustness to missing
modalities. Our proposal shows an absolute improvement of up to 10% on robust
mIoU above the most competitive baselines. Our code is available at
https://github.com/harshm121/M3L
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 05:52:50 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Maheshwari",
"Harsh",
""
],
[
"Liu",
"Yen-Cheng",
""
],
[
"Kira",
"Zsolt",
""
]
] | TITLE: Missing Modality Robustness in Semi-Supervised Multi-Modal Semantic
Segmentation
ABSTRACT: Using multiple spatial modalities has been proven helpful in improving
semantic segmentation performance. However, there are several real-world
challenges that have yet to be addressed: (a) improving label efficiency and
(b) enhancing robustness in realistic scenarios where modalities are missing at
the test time. To address these challenges, we first propose a simple yet
efficient multi-modal fusion mechanism Linear Fusion, that performs better than
the state-of-the-art multi-modal models even with limited supervision. Second,
we propose M3L: Multi-modal Teacher for Masked Modality Learning, a
semi-supervised framework that not only improves the multi-modal performance
but also makes the model robust to the realistic missing modality scenario
using unlabeled data. We create the first benchmark for semi-supervised
multi-modal semantic segmentation and also report the robustness to missing
modalities. Our proposal shows an absolute improvement of up to 10% on robust
mIoU above the most competitive baselines. Our code is available at
https://github.com/harshm121/M3L
| no_new_dataset | 0.711017 |
2304.10758 | Tanmoy Dam | Md Rasel Sarkar, Sreenatha G. Anavatti, Tanmoy Dam, Mahardhika
Pratama, and Berlian Al Kindhi | Enhancing Wind Power Forecast Precision via Multi-head Attention
Transformer: An Investigation on Single-step and Multi-step Forecasting | This paper is accepted in IJCNN23 | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by-sa/4.0/ | The main objective of this study is to propose an enhanced wind power
forecasting (EWPF) transformer model for handling power grid operations and
boosting power market competition. It helps reliable large-scale integration of
wind power relies in large part on accurate wind power forecasting (WPF). The
proposed model is evaluated for single-step and multi-step WPF, and compared
with gated recurrent unit (GRU) and long short-term memory (LSTM) models on a
wind power dataset. The results of the study indicate that the proposed EWPF
transformer model outperforms conventional recurrent neural network (RNN)
models in terms of time-series forecasting accuracy. In particular, the results
reveal a minimum performance improvement of 5% and a maximum of 20% compared to
LSTM and GRU. These results indicate that the EWPF transformer model provides a
promising alternative for wind power forecasting and has the potential to
significantly improve the precision of WPF. The findings of this study have
implications for energy producers and researchers in the field of WPF.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 06:00:50 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Sarkar",
"Md Rasel",
""
],
[
"Anavatti",
"Sreenatha G.",
""
],
[
"Dam",
"Tanmoy",
""
],
[
"Pratama",
"Mahardhika",
""
],
[
"Kindhi",
"Berlian Al",
""
]
] | TITLE: Enhancing Wind Power Forecast Precision via Multi-head Attention
Transformer: An Investigation on Single-step and Multi-step Forecasting
ABSTRACT: The main objective of this study is to propose an enhanced wind power
forecasting (EWPF) transformer model for handling power grid operations and
boosting power market competition. It helps reliable large-scale integration of
wind power relies in large part on accurate wind power forecasting (WPF). The
proposed model is evaluated for single-step and multi-step WPF, and compared
with gated recurrent unit (GRU) and long short-term memory (LSTM) models on a
wind power dataset. The results of the study indicate that the proposed EWPF
transformer model outperforms conventional recurrent neural network (RNN)
models in terms of time-series forecasting accuracy. In particular, the results
reveal a minimum performance improvement of 5% and a maximum of 20% compared to
LSTM and GRU. These results indicate that the EWPF transformer model provides a
promising alternative for wind power forecasting and has the potential to
significantly improve the precision of WPF. The findings of this study have
implications for energy producers and researchers in the field of WPF.
| no_new_dataset | 0.710038 |
2304.10759 | Chuwei Luo | Chuwei Luo, Changxu Cheng, Qi Zheng, Cong Yao | GeoLayoutLM: Geometric Pre-training for Visual Information Extraction | CVPR 2023 Highlight | null | null | null | cs.CV cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual information extraction (VIE) plays an important role in Document
Intelligence. Generally, it is divided into two tasks: semantic entity
recognition (SER) and relation extraction (RE). Recently, pre-trained models
for documents have achieved substantial progress in VIE, particularly in SER.
However, most of the existing models learn the geometric representation in an
implicit way, which has been found insufficient for the RE task since geometric
information is especially crucial for RE. Moreover, we reveal another factor
that limits the performance of RE lies in the objective gap between the
pre-training phase and the fine-tuning phase for RE. To tackle these issues, we
propose in this paper a multi-modal framework, named GeoLayoutLM, for VIE.
GeoLayoutLM explicitly models the geometric relations in pre-training, which we
call geometric pre-training. Geometric pre-training is achieved by three
specially designed geometry-related pre-training tasks. Additionally, novel
relation heads, which are pre-trained by the geometric pre-training tasks and
fine-tuned for RE, are elaborately designed to enrich and enhance the feature
representation. According to extensive experiments on standard VIE benchmarks,
GeoLayoutLM achieves highly competitive scores in the SER task and
significantly outperforms the previous state-of-the-arts for RE (\eg, the F1
score of RE on FUNSD is boosted from 80.35\% to 89.45\%). The code and models
are publicly available at
https://github.com/AlibabaResearch/AdvancedLiterateMachinery/tree/main/DocumentUnderstanding/GeoLayoutLM
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 06:02:14 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Luo",
"Chuwei",
""
],
[
"Cheng",
"Changxu",
""
],
[
"Zheng",
"Qi",
""
],
[
"Yao",
"Cong",
""
]
] | TITLE: GeoLayoutLM: Geometric Pre-training for Visual Information Extraction
ABSTRACT: Visual information extraction (VIE) plays an important role in Document
Intelligence. Generally, it is divided into two tasks: semantic entity
recognition (SER) and relation extraction (RE). Recently, pre-trained models
for documents have achieved substantial progress in VIE, particularly in SER.
However, most of the existing models learn the geometric representation in an
implicit way, which has been found insufficient for the RE task since geometric
information is especially crucial for RE. Moreover, we reveal another factor
that limits the performance of RE lies in the objective gap between the
pre-training phase and the fine-tuning phase for RE. To tackle these issues, we
propose in this paper a multi-modal framework, named GeoLayoutLM, for VIE.
GeoLayoutLM explicitly models the geometric relations in pre-training, which we
call geometric pre-training. Geometric pre-training is achieved by three
specially designed geometry-related pre-training tasks. Additionally, novel
relation heads, which are pre-trained by the geometric pre-training tasks and
fine-tuned for RE, are elaborately designed to enrich and enhance the feature
representation. According to extensive experiments on standard VIE benchmarks,
GeoLayoutLM achieves highly competitive scores in the SER task and
significantly outperforms the previous state-of-the-arts for RE (\eg, the F1
score of RE on FUNSD is boosted from 80.35\% to 89.45\%). The code and models
are publicly available at
https://github.com/AlibabaResearch/AdvancedLiterateMachinery/tree/main/DocumentUnderstanding/GeoLayoutLM
| no_new_dataset | 0.710603 |
2304.10764 | Pengfei Fang | Pengfei Fang, Mehrtash Harandi, Trung Le, Dinh Phung | Hyperbolic Geometry in Computer Vision: A Survey | First survey paper for the hyperbolic geometry in CV applications | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hyperbolic geometry, a Riemannian manifold endowed with constant sectional
negative curvature, has been considered an alternative embedding space in many
learning scenarios, \eg, natural language processing, graph learning, \etc, as
a result of its intriguing property of encoding the data's hierarchical
structure (like irregular graph or tree-likeness data). Recent studies prove
that such data hierarchy also exists in the visual dataset, and investigate the
successful practice of hyperbolic geometry in the computer vision (CV) regime,
ranging from the classical image classification to advanced model adaptation
learning. This paper presents the first and most up-to-date literature review
of hyperbolic spaces for CV applications. To this end, we first introduce the
background of hyperbolic geometry, followed by a comprehensive investigation of
algorithms, with geometric prior of hyperbolic space, in the context of visual
applications. We also conclude this manuscript and identify possible future
directions.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 06:22:16 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Fang",
"Pengfei",
""
],
[
"Harandi",
"Mehrtash",
""
],
[
"Le",
"Trung",
""
],
[
"Phung",
"Dinh",
""
]
] | TITLE: Hyperbolic Geometry in Computer Vision: A Survey
ABSTRACT: Hyperbolic geometry, a Riemannian manifold endowed with constant sectional
negative curvature, has been considered an alternative embedding space in many
learning scenarios, \eg, natural language processing, graph learning, \etc, as
a result of its intriguing property of encoding the data's hierarchical
structure (like irregular graph or tree-likeness data). Recent studies prove
that such data hierarchy also exists in the visual dataset, and investigate the
successful practice of hyperbolic geometry in the computer vision (CV) regime,
ranging from the classical image classification to advanced model adaptation
learning. This paper presents the first and most up-to-date literature review
of hyperbolic spaces for CV applications. To this end, we first introduce the
background of hyperbolic geometry, followed by a comprehensive investigation of
algorithms, with geometric prior of hyperbolic space, in the context of visual
applications. We also conclude this manuscript and identify possible future
directions.
| no_new_dataset | 0.709988 |
2304.10767 | Lisa Bonheme | Lisa Bonheme, Marek Grzes | How good are variational autoencoders at transfer learning? | Extended version of arXiv:2205.08399 with further experiments on
transfer learning | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Variational autoencoders (VAEs) are used for transfer learning across various
research domains such as music generation or medical image analysis. However,
there is no principled way to assess before transfer which components to
retrain or whether transfer learning is likely to help on a target task. We
propose to explore this question through the lens of representational
similarity. Specifically, using Centred Kernel Alignment (CKA) to evaluate the
similarity of VAEs trained on different datasets, we show that encoders'
representations are generic but decoders' specific. Based on these insights, we
discuss the implications for selecting which components of a VAE to retrain and
propose a method to visually assess whether transfer learning is likely to help
on classification tasks.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 06:32:32 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Bonheme",
"Lisa",
""
],
[
"Grzes",
"Marek",
""
]
] | TITLE: How good are variational autoencoders at transfer learning?
ABSTRACT: Variational autoencoders (VAEs) are used for transfer learning across various
research domains such as music generation or medical image analysis. However,
there is no principled way to assess before transfer which components to
retrain or whether transfer learning is likely to help on a target task. We
propose to explore this question through the lens of representational
similarity. Specifically, using Centred Kernel Alignment (CKA) to evaluate the
similarity of VAEs trained on different datasets, we show that encoders'
representations are generic but decoders' specific. Based on these insights, we
discuss the implications for selecting which components of a VAE to retrain and
propose a method to visually assess whether transfer learning is likely to help
on classification tasks.
| no_new_dataset | 0.709208 |
2304.10768 | Yongho Yoon | Yongho Yoon, Woosuk Lee, Kwangkeun Yi | Inductive Program Synthesis via Iterative Forward-Backward Abstract
Interpretation | null | null | null | null | cs.PL | http://creativecommons.org/licenses/by/4.0/ | A key challenge in example-based program synthesis is the gigantic search
space of programs. To address this challenge, various work proposed to use
abstract interpretation to prune the search space. However, most of existing
approaches have focused only on forward abstract interpretation, and thus
cannot fully exploit the power of abstract interpretation. In this paper, we
propose a novel approach to inductive program synthesis via iterative
forward-backward abstract interpretation. The forward abstract interpretation
computes possible outputs of a program given inputs, while the backward
abstract interpretation computes possible inputs of a program given outputs. By
iteratively performing the two abstract interpretations in an alternating
fashion, we can effectively determine if any completion of each partial program
as a candidate can satisfy the input-output examples. We apply our approach to
a standard formulation, syntax-guided synthesis (SyGuS), thereby supporting a
wide range of inductive synthesis tasks. We have implemented our approach and
evaluated it on a set of benchmarks from the prior work. The experimental
results show that our approach significantly outperforms the state-of-the-art
approaches thanks to the sophisticated abstract interpretation techniques.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 06:33:02 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Yoon",
"Yongho",
""
],
[
"Lee",
"Woosuk",
""
],
[
"Yi",
"Kwangkeun",
""
]
] | TITLE: Inductive Program Synthesis via Iterative Forward-Backward Abstract
Interpretation
ABSTRACT: A key challenge in example-based program synthesis is the gigantic search
space of programs. To address this challenge, various work proposed to use
abstract interpretation to prune the search space. However, most of existing
approaches have focused only on forward abstract interpretation, and thus
cannot fully exploit the power of abstract interpretation. In this paper, we
propose a novel approach to inductive program synthesis via iterative
forward-backward abstract interpretation. The forward abstract interpretation
computes possible outputs of a program given inputs, while the backward
abstract interpretation computes possible inputs of a program given outputs. By
iteratively performing the two abstract interpretations in an alternating
fashion, we can effectively determine if any completion of each partial program
as a candidate can satisfy the input-output examples. We apply our approach to
a standard formulation, syntax-guided synthesis (SyGuS), thereby supporting a
wide range of inductive synthesis tasks. We have implemented our approach and
evaluated it on a set of benchmarks from the prior work. The experimental
results show that our approach significantly outperforms the state-of-the-art
approaches thanks to the sophisticated abstract interpretation techniques.
| no_new_dataset | 0.708629 |
2304.10782 | Krishan Rana Mr | Krishan Rana, Andrew Melnik and Niko S\"underhauf | Contrastive Language, Action, and State Pre-training for Robot Learning | null | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | In this paper, we introduce a method for unifying language, action, and state
information in a shared embedding space to facilitate a range of downstream
tasks in robot learning. Our method, Contrastive Language, Action, and State
Pre-training (CLASP), extends the CLIP formulation by incorporating
distributional learning, capturing the inherent complexities and one-to-many
relationships in behaviour-text alignment. By employing distributional outputs
for both text and behaviour encoders, our model effectively associates diverse
textual commands with a single behaviour and vice-versa. We demonstrate the
utility of our method for the following downstream tasks: zero-shot
text-behaviour retrieval, captioning unseen robot behaviours, and learning a
behaviour prior for language-conditioned reinforcement learning. Our
distributional encoders exhibit superior retrieval and captioning performance
on unseen datasets, and the ability to generate meaningful exploratory
behaviours from textual commands, capturing the intricate relationships between
language, action, and state. This work represents an initial step towards
developing a unified pre-trained model for robotics, with the potential to
generalise to a broad range of downstream tasks.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 07:19:33 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Rana",
"Krishan",
""
],
[
"Melnik",
"Andrew",
""
],
[
"Sünderhauf",
"Niko",
""
]
] | TITLE: Contrastive Language, Action, and State Pre-training for Robot Learning
ABSTRACT: In this paper, we introduce a method for unifying language, action, and state
information in a shared embedding space to facilitate a range of downstream
tasks in robot learning. Our method, Contrastive Language, Action, and State
Pre-training (CLASP), extends the CLIP formulation by incorporating
distributional learning, capturing the inherent complexities and one-to-many
relationships in behaviour-text alignment. By employing distributional outputs
for both text and behaviour encoders, our model effectively associates diverse
textual commands with a single behaviour and vice-versa. We demonstrate the
utility of our method for the following downstream tasks: zero-shot
text-behaviour retrieval, captioning unseen robot behaviours, and learning a
behaviour prior for language-conditioned reinforcement learning. Our
distributional encoders exhibit superior retrieval and captioning performance
on unseen datasets, and the ability to generate meaningful exploratory
behaviours from textual commands, capturing the intricate relationships between
language, action, and state. This work represents an initial step towards
developing a unified pre-trained model for robotics, with the potential to
generalise to a broad range of downstream tasks.
| no_new_dataset | 0.710641 |
2304.10786 | Nouhaila Innan | Nouhaila Innan and Muhammad Al-Zafar Khan | Classical-to-Quantum Sequence Encoding in Genomics | 58 pages, 14 figures | null | null | null | quant-ph cs.LG | http://creativecommons.org/licenses/by/4.0/ | DNA sequencing allows for the determination of the genetic code of an
organism, and therefore is an indispensable tool that has applications in
Medicine, Life Sciences, Evolutionary Biology, Food Sciences and Technology,
and Agriculture. In this paper, we present several novel methods of performing
classical-to-quantum data encoding inspired by various mathematical fields, and
we demonstrate these ideas within Bioinformatics. In particular, we introduce
algorithms that draw inspiration from diverse fields such as Electrical and
Electronic Engineering, Information Theory, Differential Geometry, and Neural
Network architectures. We provide a complete overview of the existing data
encoding schemes and show how to use them in Genomics. The algorithms provided
utilise lossless compression, wavelet-based encoding, and information entropy.
Moreover, we propose a contemporary method for testing encoded DNA sequences
using Quantum Boltzmann Machines. To evaluate the effectiveness of our
algorithms, we discuss a potential dataset that serves as a sandbox environment
for testing against real-world scenarios. Our research contributes to
developing classical-to-quantum data encoding methods in the science of
Bioinformatics by introducing innovative algorithms that utilise diverse fields
and advanced techniques. Our findings offer insights into the potential of
Quantum Computing in Bioinformatics and have implications for future research
in this area.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 07:35:49 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Innan",
"Nouhaila",
""
],
[
"Khan",
"Muhammad Al-Zafar",
""
]
] | TITLE: Classical-to-Quantum Sequence Encoding in Genomics
ABSTRACT: DNA sequencing allows for the determination of the genetic code of an
organism, and therefore is an indispensable tool that has applications in
Medicine, Life Sciences, Evolutionary Biology, Food Sciences and Technology,
and Agriculture. In this paper, we present several novel methods of performing
classical-to-quantum data encoding inspired by various mathematical fields, and
we demonstrate these ideas within Bioinformatics. In particular, we introduce
algorithms that draw inspiration from diverse fields such as Electrical and
Electronic Engineering, Information Theory, Differential Geometry, and Neural
Network architectures. We provide a complete overview of the existing data
encoding schemes and show how to use them in Genomics. The algorithms provided
utilise lossless compression, wavelet-based encoding, and information entropy.
Moreover, we propose a contemporary method for testing encoded DNA sequences
using Quantum Boltzmann Machines. To evaluate the effectiveness of our
algorithms, we discuss a potential dataset that serves as a sandbox environment
for testing against real-world scenarios. Our research contributes to
developing classical-to-quantum data encoding methods in the science of
Bioinformatics by introducing innovative algorithms that utilise diverse fields
and advanced techniques. Our findings offer insights into the potential of
Quantum Computing in Bioinformatics and have implications for future research
in this area.
| no_new_dataset | 0.640158 |
2304.10805 | YongTaek Lim | Yewon Kim, YongTaek Lim, Dokyung Yoon and KyungWoo Song | RPLKG: Robust Prompt Learning with Knowledge Graph | null | null | null | null | cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Large-scale pre-trained models have been known that they are transferable,
and they generalize well on the unseen dataset. Recently, multimodal
pre-trained models such as CLIP show significant performance improvement in
diverse experiments. However, when the labeled dataset is limited, the
generalization of a new dataset or domain is still challenging. To improve the
generalization performance on few-shot learning, there have been diverse
efforts, such as prompt learning and adapter. However, the current few-shot
adaptation methods are not interpretable, and they require a high computation
cost for adaptation. In this study, we propose a new method, robust prompt
learning with knowledge graph (RPLKG). Based on the knowledge graph, we
automatically design diverse interpretable and meaningful prompt sets. Our
model obtains cached embeddings of prompt sets after one forwarding from a
large pre-trained model. After that, model optimizes the prompt selection
processes with GumbelSoftmax. In this way, our model is trained using
relatively little memory and learning time. Also, RPLKG selects the optimal
interpretable prompt automatically, depending on the dataset. In summary, RPLKG
is i) interpretable, ii) requires small computation resources, and iii) easy to
incorporate prior human knowledge. To validate the RPLKG, we provide
comprehensive experimental results on few-shot learning, domain generalization
and new class generalization setting. RPLKG shows a significant performance
improvement compared to zero-shot learning and competitive performance against
several prompt learning methods using much lower resources.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 08:22:58 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Kim",
"Yewon",
""
],
[
"Lim",
"YongTaek",
""
],
[
"Yoon",
"Dokyung",
""
],
[
"Song",
"KyungWoo",
""
]
] | TITLE: RPLKG: Robust Prompt Learning with Knowledge Graph
ABSTRACT: Large-scale pre-trained models have been known that they are transferable,
and they generalize well on the unseen dataset. Recently, multimodal
pre-trained models such as CLIP show significant performance improvement in
diverse experiments. However, when the labeled dataset is limited, the
generalization of a new dataset or domain is still challenging. To improve the
generalization performance on few-shot learning, there have been diverse
efforts, such as prompt learning and adapter. However, the current few-shot
adaptation methods are not interpretable, and they require a high computation
cost for adaptation. In this study, we propose a new method, robust prompt
learning with knowledge graph (RPLKG). Based on the knowledge graph, we
automatically design diverse interpretable and meaningful prompt sets. Our
model obtains cached embeddings of prompt sets after one forwarding from a
large pre-trained model. After that, model optimizes the prompt selection
processes with GumbelSoftmax. In this way, our model is trained using
relatively little memory and learning time. Also, RPLKG selects the optimal
interpretable prompt automatically, depending on the dataset. In summary, RPLKG
is i) interpretable, ii) requires small computation resources, and iii) easy to
incorporate prior human knowledge. To validate the RPLKG, we provide
comprehensive experimental results on few-shot learning, domain generalization
and new class generalization setting. RPLKG shows a significant performance
improvement compared to zero-shot learning and competitive performance against
several prompt learning methods using much lower resources.
| no_new_dataset | 0.707847 |
2304.10814 | Alexander Tsaregorodtsev | Alexander Tsaregorodtsev, Adrian Holzbock, Jan Strohbeck, Michael
Buchholz, and Vasileios Belagiannis | Automated Static Camera Calibration with Intelligent Vehicles | 7 pages, 3 figures, accepted for presentation at the 34th IEEE
Intelligent Vehicles Symposium (IV 2023), June 4 - June 7, 2023, Anchorage,
Alaska, United States of America | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Connected and cooperative driving requires precise calibration of the
roadside infrastructure for having a reliable perception system. To solve this
requirement in an automated manner, we present a robust extrinsic calibration
method for automated geo-referenced camera calibration. Our method requires a
calibration vehicle equipped with a combined GNSS/RTK receiver and an inertial
measurement unit (IMU) for self-localization. In order to remove any
requirements for the target's appearance and the local traffic conditions, we
propose a novel approach using hypothesis filtering. Our method does not
require any human interaction with the information recorded by both the
infrastructure and the vehicle. Furthermore, we do not limit road access for
other road users during calibration. We demonstrate the feasibility and
accuracy of our approach by evaluating our approach on synthetic datasets as
well as a real-world connected intersection, and deploying the calibration on
real infrastructure. Our source code is publicly available.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 08:50:52 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Tsaregorodtsev",
"Alexander",
""
],
[
"Holzbock",
"Adrian",
""
],
[
"Strohbeck",
"Jan",
""
],
[
"Buchholz",
"Michael",
""
],
[
"Belagiannis",
"Vasileios",
""
]
] | TITLE: Automated Static Camera Calibration with Intelligent Vehicles
ABSTRACT: Connected and cooperative driving requires precise calibration of the
roadside infrastructure for having a reliable perception system. To solve this
requirement in an automated manner, we present a robust extrinsic calibration
method for automated geo-referenced camera calibration. Our method requires a
calibration vehicle equipped with a combined GNSS/RTK receiver and an inertial
measurement unit (IMU) for self-localization. In order to remove any
requirements for the target's appearance and the local traffic conditions, we
propose a novel approach using hypothesis filtering. Our method does not
require any human interaction with the information recorded by both the
infrastructure and the vehicle. Furthermore, we do not limit road access for
other road users during calibration. We demonstrate the feasibility and
accuracy of our approach by evaluating our approach on synthetic datasets as
well as a real-world connected intersection, and deploying the calibration on
real infrastructure. Our source code is publicly available.
| no_new_dataset | 0.706152 |
2304.10824 | Weijing Chen | Weijing Chen, Linli Yao, Qin Jin | Rethinking Benchmarks for Cross-modal Image-text Retrieval | Accepted to SIGIR2023 | null | 10.1145/3539618.3591758 | null | cs.CV cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Image-text retrieval, as a fundamental and important branch of information
retrieval, has attracted extensive research attentions. The main challenge of
this task is cross-modal semantic understanding and matching. Some recent works
focus more on fine-grained cross-modal semantic matching. With the prevalence
of large scale multimodal pretraining models, several state-of-the-art models
(e.g. X-VLM) have achieved near-perfect performance on widely-used image-text
retrieval benchmarks, i.e. MSCOCO-Test-5K and Flickr30K-Test-1K. In this paper,
we review the two common benchmarks and observe that they are insufficient to
assess the true capability of models on fine-grained cross-modal semantic
matching. The reason is that a large amount of images and texts in the
benchmarks are coarse-grained. Based on the observation, we renovate the
coarse-grained images and texts in the old benchmarks and establish the
improved benchmarks called MSCOCO-FG and Flickr30K-FG. Specifically, on the
image side, we enlarge the original image pool by adopting more similar images.
On the text side, we propose a novel semi-automatic renovation approach to
refine coarse-grained sentences into finer-grained ones with little human
effort. Furthermore, we evaluate representative image-text retrieval models on
our new benchmarks to demonstrate the effectiveness of our method. We also
analyze the capability of models on fine-grained semantic comprehension through
extensive experiments. The results show that even the state-of-the-art models
have much room for improvement in fine-grained semantic understanding,
especially in distinguishing attributes of close objects in images. Our code
and improved benchmark datasets are publicly available at:
https://github.com/cwj1412/MSCOCO-Flikcr30K_FG, which we hope will inspire
further in-depth research on cross-modal retrieval.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 09:07:57 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Chen",
"Weijing",
""
],
[
"Yao",
"Linli",
""
],
[
"Jin",
"Qin",
""
]
] | TITLE: Rethinking Benchmarks for Cross-modal Image-text Retrieval
ABSTRACT: Image-text retrieval, as a fundamental and important branch of information
retrieval, has attracted extensive research attentions. The main challenge of
this task is cross-modal semantic understanding and matching. Some recent works
focus more on fine-grained cross-modal semantic matching. With the prevalence
of large scale multimodal pretraining models, several state-of-the-art models
(e.g. X-VLM) have achieved near-perfect performance on widely-used image-text
retrieval benchmarks, i.e. MSCOCO-Test-5K and Flickr30K-Test-1K. In this paper,
we review the two common benchmarks and observe that they are insufficient to
assess the true capability of models on fine-grained cross-modal semantic
matching. The reason is that a large amount of images and texts in the
benchmarks are coarse-grained. Based on the observation, we renovate the
coarse-grained images and texts in the old benchmarks and establish the
improved benchmarks called MSCOCO-FG and Flickr30K-FG. Specifically, on the
image side, we enlarge the original image pool by adopting more similar images.
On the text side, we propose a novel semi-automatic renovation approach to
refine coarse-grained sentences into finer-grained ones with little human
effort. Furthermore, we evaluate representative image-text retrieval models on
our new benchmarks to demonstrate the effectiveness of our method. We also
analyze the capability of models on fine-grained semantic comprehension through
extensive experiments. The results show that even the state-of-the-art models
have much room for improvement in fine-grained semantic understanding,
especially in distinguishing attributes of close objects in images. Our code
and improved benchmark datasets are publicly available at:
https://github.com/cwj1412/MSCOCO-Flikcr30K_FG, which we hope will inspire
further in-depth research on cross-modal retrieval.
| no_new_dataset | 0.70844 |
2304.10825 | Chengyu Zheng | Chengyu Zheng, Peng Li, Xiao-Ping Zhang, Xuequan Lu, Mingqiang Wei | Don't worry about mistakes! Glass Segmentation Network via Mistake
Correction | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recall one time when we were in an unfamiliar mall. We might mistakenly think
that there exists or does not exist a piece of glass in front of us. Such
mistakes will remind us to walk more safely and freely at the same or a similar
place next time. To absorb the human mistake correction wisdom, we propose a
novel glass segmentation network to detect transparent glass, dubbed
GlassSegNet. Motivated by this human behavior, GlassSegNet utilizes two key
stages: the identification stage (IS) and the correction stage (CS). The IS is
designed to simulate the detection procedure of human recognition for
identifying transparent glass by global context and edge information. The CS
then progressively refines the coarse prediction by correcting mistake regions
based on gained experience. Extensive experiments show clear improvements of
our GlassSegNet over thirty-four state-of-the-art methods on three benchmark
datasets.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 09:09:00 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Zheng",
"Chengyu",
""
],
[
"Li",
"Peng",
""
],
[
"Zhang",
"Xiao-Ping",
""
],
[
"Lu",
"Xuequan",
""
],
[
"Wei",
"Mingqiang",
""
]
] | TITLE: Don't worry about mistakes! Glass Segmentation Network via Mistake
Correction
ABSTRACT: Recall one time when we were in an unfamiliar mall. We might mistakenly think
that there exists or does not exist a piece of glass in front of us. Such
mistakes will remind us to walk more safely and freely at the same or a similar
place next time. To absorb the human mistake correction wisdom, we propose a
novel glass segmentation network to detect transparent glass, dubbed
GlassSegNet. Motivated by this human behavior, GlassSegNet utilizes two key
stages: the identification stage (IS) and the correction stage (CS). The IS is
designed to simulate the detection procedure of human recognition for
identifying transparent glass by global context and edge information. The CS
then progressively refines the coarse prediction by correcting mistake regions
based on gained experience. Extensive experiments show clear improvements of
our GlassSegNet over thirty-four state-of-the-art methods on three benchmark
datasets.
| no_new_dataset | 0.706994 |
2304.10827 | Niccol\`o Di Marco | Niccol\`o Di Marco, Matteo Cinelli, Shayan Alipour and Walter
Quattrociocchi | Users volatility on Reddit and Voat | null | null | null | null | physics.soc-ph cs.SI | http://creativecommons.org/licenses/by-sa/4.0/ | Social media platforms are like giant arenas where users can rely on
different content and express their opinions through likes, comments, and
shares. However, do users welcome different perspectives or only listen to
their preferred narratives? This paper examines how users explore the digital
space and allocate their attention among communities on two social networks,
Voat and Reddit. By analysing a massive dataset of about 215 million comments
posted by about 16 million users on Voat and Reddit in 2019 we find that most
users tend to explore new communities at a decreasing rate, meaning they have a
limited set of preferred groups they visit regularly. Moreover, we provide
evidence that preferred communities of users tend to cover similar topics
throughout the year. We also find that communities have a high turnover of
users, meaning that users come and go frequently showing a high volatility that
strongly departs from a null model simulating users' behaviour.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 09:11:42 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Di Marco",
"Niccolò",
""
],
[
"Cinelli",
"Matteo",
""
],
[
"Alipour",
"Shayan",
""
],
[
"Quattrociocchi",
"Walter",
""
]
] | TITLE: Users volatility on Reddit and Voat
ABSTRACT: Social media platforms are like giant arenas where users can rely on
different content and express their opinions through likes, comments, and
shares. However, do users welcome different perspectives or only listen to
their preferred narratives? This paper examines how users explore the digital
space and allocate their attention among communities on two social networks,
Voat and Reddit. By analysing a massive dataset of about 215 million comments
posted by about 16 million users on Voat and Reddit in 2019 we find that most
users tend to explore new communities at a decreasing rate, meaning they have a
limited set of preferred groups they visit regularly. Moreover, we provide
evidence that preferred communities of users tend to cover similar topics
throughout the year. We also find that communities have a high turnover of
users, meaning that users come and go frequently showing a high volatility that
strongly departs from a null model simulating users' behaviour.
| no_new_dataset | 0.709239 |
2304.10828 | Andrea Patane | Alice Doherty, Matthew Wicker, Luca Laurenti, Andrea Patane | Individual Fairness in Bayesian Neural Networks | null | null | null | null | cs.LG cs.CY | http://creativecommons.org/licenses/by/4.0/ | We study Individual Fairness (IF) for Bayesian neural networks (BNNs).
Specifically, we consider the $\epsilon$-$\delta$-individual fairness notion,
which requires that, for any pair of input points that are $\epsilon$-similar
according to a given similarity metrics, the output of the BNN is within a
given tolerance $\delta>0.$ We leverage bounds on statistical sampling over the
input space and the relationship between adversarial robustness and individual
fairness to derive a framework for the systematic estimation of
$\epsilon$-$\delta$-IF, designing Fair-FGSM and Fair-PGD as
global,fairness-aware extensions to gradient-based attacks for BNNs. We
empirically study IF of a variety of approximately inferred BNNs with different
architectures on fairness benchmarks, and compare against deterministic models
learnt using frequentist techniques. Interestingly, we find that BNNs trained
by means of approximate Bayesian inference consistently tend to be markedly
more individually fair than their deterministic counterparts.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 09:12:14 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Doherty",
"Alice",
""
],
[
"Wicker",
"Matthew",
""
],
[
"Laurenti",
"Luca",
""
],
[
"Patane",
"Andrea",
""
]
] | TITLE: Individual Fairness in Bayesian Neural Networks
ABSTRACT: We study Individual Fairness (IF) for Bayesian neural networks (BNNs).
Specifically, we consider the $\epsilon$-$\delta$-individual fairness notion,
which requires that, for any pair of input points that are $\epsilon$-similar
according to a given similarity metrics, the output of the BNN is within a
given tolerance $\delta>0.$ We leverage bounds on statistical sampling over the
input space and the relationship between adversarial robustness and individual
fairness to derive a framework for the systematic estimation of
$\epsilon$-$\delta$-IF, designing Fair-FGSM and Fair-PGD as
global,fairness-aware extensions to gradient-based attacks for BNNs. We
empirically study IF of a variety of approximately inferred BNNs with different
architectures on fairness benchmarks, and compare against deterministic models
learnt using frequentist techniques. Interestingly, we find that BNNs trained
by means of approximate Bayesian inference consistently tend to be markedly
more individually fair than their deterministic counterparts.
| no_new_dataset | 0.709044 |
2304.10831 | Yuan Cao | Yuan Cao, Di Jiang, Guanqun Hou, Fan Deng, Xinjia Chen, Qiang Yang | Learn to Cluster Faces with Better Subgraphs | null | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Face clustering can provide pseudo-labels to the massive unlabeled face data
and improve the performance of different face recognition models. The existing
clustering methods generally aggregate the features within subgraphs that are
often implemented based on a uniform threshold or a learned cutoff position.
This may reduce the recall of subgraphs and hence degrade the clustering
performance. This work proposed an efficient neighborhood-aware subgraph
adjustment method that can significantly reduce the noise and improve the
recall of the subgraphs, and hence can drive the distant nodes to converge
towards the same centers. More specifically, the proposed method consists of
two components, i.e. face embeddings enhancement using the embeddings from
neighbors, and enclosed subgraph construction of node pairs for structural
information extraction. The embeddings are combined to predict the linkage
probabilities for all node pairs to replace the cosine similarities to produce
new subgraphs that can be further used for aggregation of GCNs or other
clustering methods. The proposed method is validated through extensive
experiments against a range of clustering solutions using three benchmark
datasets and numerical results confirm that it outperforms the SOTA solutions
in terms of generalization capability.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 09:18:55 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Cao",
"Yuan",
""
],
[
"Jiang",
"Di",
""
],
[
"Hou",
"Guanqun",
""
],
[
"Deng",
"Fan",
""
],
[
"Chen",
"Xinjia",
""
],
[
"Yang",
"Qiang",
""
]
] | TITLE: Learn to Cluster Faces with Better Subgraphs
ABSTRACT: Face clustering can provide pseudo-labels to the massive unlabeled face data
and improve the performance of different face recognition models. The existing
clustering methods generally aggregate the features within subgraphs that are
often implemented based on a uniform threshold or a learned cutoff position.
This may reduce the recall of subgraphs and hence degrade the clustering
performance. This work proposed an efficient neighborhood-aware subgraph
adjustment method that can significantly reduce the noise and improve the
recall of the subgraphs, and hence can drive the distant nodes to converge
towards the same centers. More specifically, the proposed method consists of
two components, i.e. face embeddings enhancement using the embeddings from
neighbors, and enclosed subgraph construction of node pairs for structural
information extraction. The embeddings are combined to predict the linkage
probabilities for all node pairs to replace the cosine similarities to produce
new subgraphs that can be further used for aggregation of GCNs or other
clustering methods. The proposed method is validated through extensive
experiments against a range of clustering solutions using three benchmark
datasets and numerical results confirm that it outperforms the SOTA solutions
in terms of generalization capability.
| no_new_dataset | 0.710672 |
2304.10844 | Ru Peng | Ru Peng, Yawen Zeng, Junbo Zhao | Better Sign Language Translation with Monolingual Data | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sign language translation (SLT) systems, which are often decomposed into
video-to-gloss (V2G) recognition and gloss-to-text (G2T) translation through
the pivot gloss, heavily relies on the availability of large-scale parallel G2T
pairs. However, the manual annotation of pivot gloss, which is a sequence of
transcribed written-language words in the order in which they are signed,
further exacerbates the scarcity of data for SLT. To address this issue, this
paper proposes a simple and efficient rule transformation method to transcribe
the large-scale target monolingual data into its pseudo glosses automatically
for enhancing the SLT translation. Empirical results show that the proposed
approach can significantly improve the performance of SLT, especially achieving
state-of-the-art results on two SLT benchmark datasets PHEONIX-WEATHER 2014T
and ASLG-PC12. Our code has been released at:
https://github.com/pengr/Mono\_SLT.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 09:39:54 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Peng",
"Ru",
""
],
[
"Zeng",
"Yawen",
""
],
[
"Zhao",
"Junbo",
""
]
] | TITLE: Better Sign Language Translation with Monolingual Data
ABSTRACT: Sign language translation (SLT) systems, which are often decomposed into
video-to-gloss (V2G) recognition and gloss-to-text (G2T) translation through
the pivot gloss, heavily relies on the availability of large-scale parallel G2T
pairs. However, the manual annotation of pivot gloss, which is a sequence of
transcribed written-language words in the order in which they are signed,
further exacerbates the scarcity of data for SLT. To address this issue, this
paper proposes a simple and efficient rule transformation method to transcribe
the large-scale target monolingual data into its pseudo glosses automatically
for enhancing the SLT translation. Empirical results show that the proposed
approach can significantly improve the performance of SLT, especially achieving
state-of-the-art results on two SLT benchmark datasets PHEONIX-WEATHER 2014T
and ASLG-PC12. Our code has been released at:
https://github.com/pengr/Mono\_SLT.
| no_new_dataset | 0.711224 |
2304.10854 | Zhengcheng Shen | Zhengcheng Shen, Yi Gao, Linh K\"astner, Jens Lambrecht | HabitatDyn Dataset: Dynamic Object Detection to Kinematics Estimation | The paper is under review | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The advancement of computer vision and machine learning has made datasets a
crucial element for further research and applications. However, the creation
and development of robots with advanced recognition capabilities are hindered
by the lack of appropriate datasets. Existing image or video processing
datasets are unable to accurately depict observations from a moving robot, and
they do not contain the kinematics information necessary for robotic tasks.
Synthetic data, on the other hand, are cost-effective to create and offer
greater flexibility for adapting to various applications. Hence, they are
widely utilized in both research and industry. In this paper, we propose the
dataset HabitatDyn, which contains both synthetic RGB videos, semantic labels,
and depth information, as well as kinetics information. HabitatDyn was created
from the perspective of a mobile robot with a moving camera, and contains 30
scenes featuring six different types of moving objects with varying velocities.
To demonstrate the usability of our dataset, two existing algorithms are used
for evaluation and an approach to estimate the distance between the object and
camera is implemented based on these segmentation methods and evaluated through
the dataset. With the availability of this dataset, we aspire to foster further
advancements in the field of mobile robotics, leading to more capable and
intelligent robots that can navigate and interact with their environments more
effectively. The code is publicly available at
https://github.com/ignc-research/HabitatDyn.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 09:57:35 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Shen",
"Zhengcheng",
""
],
[
"Gao",
"Yi",
""
],
[
"Kästner",
"Linh",
""
],
[
"Lambrecht",
"Jens",
""
]
] | TITLE: HabitatDyn Dataset: Dynamic Object Detection to Kinematics Estimation
ABSTRACT: The advancement of computer vision and machine learning has made datasets a
crucial element for further research and applications. However, the creation
and development of robots with advanced recognition capabilities are hindered
by the lack of appropriate datasets. Existing image or video processing
datasets are unable to accurately depict observations from a moving robot, and
they do not contain the kinematics information necessary for robotic tasks.
Synthetic data, on the other hand, are cost-effective to create and offer
greater flexibility for adapting to various applications. Hence, they are
widely utilized in both research and industry. In this paper, we propose the
dataset HabitatDyn, which contains both synthetic RGB videos, semantic labels,
and depth information, as well as kinetics information. HabitatDyn was created
from the perspective of a mobile robot with a moving camera, and contains 30
scenes featuring six different types of moving objects with varying velocities.
To demonstrate the usability of our dataset, two existing algorithms are used
for evaluation and an approach to estimate the distance between the object and
camera is implemented based on these segmentation methods and evaluated through
the dataset. With the availability of this dataset, we aspire to foster further
advancements in the field of mobile robotics, leading to more capable and
intelligent robots that can navigate and interact with their environments more
effectively. The code is publicly available at
https://github.com/ignc-research/HabitatDyn.
| new_dataset | 0.718826 |
2304.10869 | Mohan Li | Mohan Li and Rama Doddipatla | Non-autoregressive End-to-end Approaches for Joint Automatic Speech
Recognition and Spoken Language Understanding | 8 pages, 1 figure, accepted at IEEE SLT2023 | null | null | null | eess.AS cs.SD | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This paper presents the use of non-autoregressive (NAR) approaches for joint
automatic speech recognition (ASR) and spoken language understanding (SLU)
tasks. The proposed NAR systems employ a Conformer encoder that applies
connectionist temporal classification (CTC) to transcribe the speech utterance
into raw ASR hypotheses, which are further refined with a bidirectional encoder
representations from Transformers (BERT)-like decoder. In the meantime, the
intent and slot labels of the utterance are predicted simultaneously using the
same decoder. Both Mask-CTC and self-conditioned CTC (SC-CTC) approaches are
explored for this study. Experiments conducted on the SLURP dataset show that
the proposed SC-Mask-CTC NAR system achieves 3.7% and 3.2% absolute gains in
SLU metrics and a competitive level of ASR accuracy, when compared to a
Conformer-Transformer based autoregressive (AR) model. Additionally, the NAR
systems achieve 6x faster decoding speed than the AR baseline.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 10:31:44 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Li",
"Mohan",
""
],
[
"Doddipatla",
"Rama",
""
]
] | TITLE: Non-autoregressive End-to-end Approaches for Joint Automatic Speech
Recognition and Spoken Language Understanding
ABSTRACT: This paper presents the use of non-autoregressive (NAR) approaches for joint
automatic speech recognition (ASR) and spoken language understanding (SLU)
tasks. The proposed NAR systems employ a Conformer encoder that applies
connectionist temporal classification (CTC) to transcribe the speech utterance
into raw ASR hypotheses, which are further refined with a bidirectional encoder
representations from Transformers (BERT)-like decoder. In the meantime, the
intent and slot labels of the utterance are predicted simultaneously using the
same decoder. Both Mask-CTC and self-conditioned CTC (SC-CTC) approaches are
explored for this study. Experiments conducted on the SLURP dataset show that
the proposed SC-Mask-CTC NAR system achieves 3.7% and 3.2% absolute gains in
SLU metrics and a competitive level of ASR accuracy, when compared to a
Conformer-Transformer based autoregressive (AR) model. Additionally, the NAR
systems achieve 6x faster decoding speed than the AR baseline.
| no_new_dataset | 0.710189 |
2304.10882 | Aiqing Zhu | Jiawei Zhang, Aiqing Zhu, Feng Ji, Chang Lin and Yifa Tang | Effective Numerical Simulations of Synchronous Generator System | null | null | null | null | math.NA cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Synchronous generator system is a complicated dynamical system for energy
transmission, which plays an important role in modern industrial production. In
this article, we propose some predictor-corrector methods and
structure-preserving methods for a generator system based on the first
benchmark model of subsynchronous resonance, among which the
structure-preserving methods preserve a Dirac structure associated with the
so-called port-Hamiltonian descriptor systems. To illustrate this, the
simplified generator system in the form of index-1 differential-algebraic
equations has been derived. Our analyses provide the global error estimates for
a special class of structure-preserving methods called Gauss methods, which
guarantee their superior performance over the PSCAD/EMTDC and the
predictor-corrector methods in terms of computational stability. Numerical
simulations are implemented to verify the effectiveness and advantages of our
methods.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 10:51:00 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Zhang",
"Jiawei",
""
],
[
"Zhu",
"Aiqing",
""
],
[
"Ji",
"Feng",
""
],
[
"Lin",
"Chang",
""
],
[
"Tang",
"Yifa",
""
]
] | TITLE: Effective Numerical Simulations of Synchronous Generator System
ABSTRACT: Synchronous generator system is a complicated dynamical system for energy
transmission, which plays an important role in modern industrial production. In
this article, we propose some predictor-corrector methods and
structure-preserving methods for a generator system based on the first
benchmark model of subsynchronous resonance, among which the
structure-preserving methods preserve a Dirac structure associated with the
so-called port-Hamiltonian descriptor systems. To illustrate this, the
simplified generator system in the form of index-1 differential-algebraic
equations has been derived. Our analyses provide the global error estimates for
a special class of structure-preserving methods called Gauss methods, which
guarantee their superior performance over the PSCAD/EMTDC and the
predictor-corrector methods in terms of computational stability. Numerical
simulations are implemented to verify the effectiveness and advantages of our
methods.
| no_new_dataset | 0.709453 |
2304.10891 | Xi Chen | Juan Zhong, Zheng Liu, Xi Chen | Transformer-based models and hardware acceleration analysis in
autonomous driving: A survey | null | null | null | null | cs.LG cs.AI cs.CV cs.RO cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Transformer architectures have exhibited promising performance in various
autonomous driving applications in recent years. On the other hand, its
dedicated hardware acceleration on portable computational platforms has become
the next critical step for practical deployment in real autonomous vehicles.
This survey paper provides a comprehensive overview, benchmark, and analysis of
Transformer-based models specifically tailored for autonomous driving tasks
such as lane detection, segmentation, tracking, planning, and decision-making.
We review different architectures for organizing Transformer inputs and
outputs, such as encoder-decoder and encoder-only structures, and explore their
respective advantages and disadvantages. Furthermore, we discuss
Transformer-related operators and their hardware acceleration schemes in depth,
taking into account key factors such as quantization and runtime. We
specifically illustrate the operator level comparison between layers from
convolutional neural network, Swin-Transformer, and Transformer with 4D
encoder. The paper also highlights the challenges, trends, and current insights
in Transformer-based models, addressing their hardware deployment and
acceleration issues within the context of long-term autonomous driving
applications.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 11:15:31 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Zhong",
"Juan",
""
],
[
"Liu",
"Zheng",
""
],
[
"Chen",
"Xi",
""
]
] | TITLE: Transformer-based models and hardware acceleration analysis in
autonomous driving: A survey
ABSTRACT: Transformer architectures have exhibited promising performance in various
autonomous driving applications in recent years. On the other hand, its
dedicated hardware acceleration on portable computational platforms has become
the next critical step for practical deployment in real autonomous vehicles.
This survey paper provides a comprehensive overview, benchmark, and analysis of
Transformer-based models specifically tailored for autonomous driving tasks
such as lane detection, segmentation, tracking, planning, and decision-making.
We review different architectures for organizing Transformer inputs and
outputs, such as encoder-decoder and encoder-only structures, and explore their
respective advantages and disadvantages. Furthermore, we discuss
Transformer-related operators and their hardware acceleration schemes in depth,
taking into account key factors such as quantization and runtime. We
specifically illustrate the operator level comparison between layers from
convolutional neural network, Swin-Transformer, and Transformer with 4D
encoder. The paper also highlights the challenges, trends, and current insights
in Transformer-based models, addressing their hardware deployment and
acceleration issues within the context of long-term autonomous driving
applications.
| no_new_dataset | 0.708042 |
2304.10893 | Runwei Guan | Runwei Guan, Ka Lok Man, Feifan Chen, Shanliang Yao, Rongsheng Hu,
Xiaohui Zhu, Jeremy Smith, Eng Gee Lim and Yutao Yue | FindVehicle and VehicleFinder: A NER dataset for natural language-based
vehicle retrieval and a keyword-based cross-modal vehicle retrieval system | null | null | null | null | cs.CV cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Natural language (NL) based vehicle retrieval is a task aiming to retrieve a
vehicle that is most consistent with a given NL query from among all candidate
vehicles. Because NL query can be easily obtained, such a task has a promising
prospect in building an interactive intelligent traffic system (ITS). Current
solutions mainly focus on extracting both text and image features and mapping
them to the same latent space to compare the similarity. However, existing
methods usually use dependency analysis or semantic role-labelling techniques
to find keywords related to vehicle attributes. These techniques may require a
lot of pre-processing and post-processing work, and also suffer from extracting
the wrong keyword when the NL query is complex. To tackle these problems and
simplify, we borrow the idea from named entity recognition (NER) and construct
FindVehicle, a NER dataset in the traffic domain. It has 42.3k labelled NL
descriptions of vehicle tracks, containing information such as the location,
orientation, type and colour of the vehicle. FindVehicle also adopts both
overlapping entities and fine-grained entities to meet further requirements. To
verify its effectiveness, we propose a baseline NL-based vehicle retrieval
model called VehicleFinder. Our experiment shows that by using text encoders
pre-trained by FindVehicle, VehicleFinder achieves 87.7\% precision and 89.4\%
recall when retrieving a target vehicle by text command on our homemade dataset
based on UA-DETRAC. The time cost of VehicleFinder is 279.35 ms on one ARM v8.2
CPU and 93.72 ms on one RTX A4000 GPU, which is much faster than the
Transformer-based system. The dataset is open-source via the link
https://github.com/GuanRunwei/FindVehicle, and the implementation can be found
via the link https://github.com/GuanRunwei/VehicleFinder-CTIM.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 11:20:23 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Guan",
"Runwei",
""
],
[
"Man",
"Ka Lok",
""
],
[
"Chen",
"Feifan",
""
],
[
"Yao",
"Shanliang",
""
],
[
"Hu",
"Rongsheng",
""
],
[
"Zhu",
"Xiaohui",
""
],
[
"Smith",
"Jeremy",
""
],
[
"Lim",
"Eng Gee",
""
],
[
"Yue",
"Yutao",
""
]
] | TITLE: FindVehicle and VehicleFinder: A NER dataset for natural language-based
vehicle retrieval and a keyword-based cross-modal vehicle retrieval system
ABSTRACT: Natural language (NL) based vehicle retrieval is a task aiming to retrieve a
vehicle that is most consistent with a given NL query from among all candidate
vehicles. Because NL query can be easily obtained, such a task has a promising
prospect in building an interactive intelligent traffic system (ITS). Current
solutions mainly focus on extracting both text and image features and mapping
them to the same latent space to compare the similarity. However, existing
methods usually use dependency analysis or semantic role-labelling techniques
to find keywords related to vehicle attributes. These techniques may require a
lot of pre-processing and post-processing work, and also suffer from extracting
the wrong keyword when the NL query is complex. To tackle these problems and
simplify, we borrow the idea from named entity recognition (NER) and construct
FindVehicle, a NER dataset in the traffic domain. It has 42.3k labelled NL
descriptions of vehicle tracks, containing information such as the location,
orientation, type and colour of the vehicle. FindVehicle also adopts both
overlapping entities and fine-grained entities to meet further requirements. To
verify its effectiveness, we propose a baseline NL-based vehicle retrieval
model called VehicleFinder. Our experiment shows that by using text encoders
pre-trained by FindVehicle, VehicleFinder achieves 87.7\% precision and 89.4\%
recall when retrieving a target vehicle by text command on our homemade dataset
based on UA-DETRAC. The time cost of VehicleFinder is 279.35 ms on one ARM v8.2
CPU and 93.72 ms on one RTX A4000 GPU, which is much faster than the
Transformer-based system. The dataset is open-source via the link
https://github.com/GuanRunwei/FindVehicle, and the implementation can be found
via the link https://github.com/GuanRunwei/VehicleFinder-CTIM.
| no_new_dataset | 0.702785 |
2304.10909 | Joakim Edin | Joakim Edin, Alexander Junge, Jakob D. Havtorn, Lasse Borgholt, Maria
Maistro, Tuukka Ruotsalo, Lars Maal{\o}e | Automated Medical Coding on MIMIC-III and MIMIC-IV: A Critical Review
and Replicability Study | 11 pages, 6 figures, to be published in Proceedings of the 46th
International ACM SIGIR Conference on Research and Development in Information
Retrieval (SIGIR '23), July 23--27, 2023, Taipei, Taiwan | null | 10.1145/3539618.3591918 | null | cs.LG cs.IR | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Medical coding is the task of assigning medical codes to clinical free-text
documentation. Healthcare professionals manually assign such codes to track
patient diagnoses and treatments. Automated medical coding can considerably
alleviate this administrative burden. In this paper, we reproduce, compare, and
analyze state-of-the-art automated medical coding machine learning models. We
show that several models underperform due to weak configurations, poorly
sampled train-test splits, and insufficient evaluation. In previous work, the
macro F1 score has been calculated sub-optimally, and our correction doubles
it. We contribute a revised model comparison using stratified sampling and
identical experimental setups, including hyperparameters and decision boundary
tuning. We analyze prediction errors to validate and falsify assumptions of
previous works. The analysis confirms that all models struggle with rare codes,
while long documents only have a negligible impact. Finally, we present the
first comprehensive results on the newly released MIMIC-IV dataset using the
reproduced models. We release our code, model parameters, and new MIMIC-III and
MIMIC-IV training and evaluation pipelines to accommodate fair future
comparisons.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 11:54:44 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Edin",
"Joakim",
""
],
[
"Junge",
"Alexander",
""
],
[
"Havtorn",
"Jakob D.",
""
],
[
"Borgholt",
"Lasse",
""
],
[
"Maistro",
"Maria",
""
],
[
"Ruotsalo",
"Tuukka",
""
],
[
"Maaløe",
"Lars",
""
]
] | TITLE: Automated Medical Coding on MIMIC-III and MIMIC-IV: A Critical Review
and Replicability Study
ABSTRACT: Medical coding is the task of assigning medical codes to clinical free-text
documentation. Healthcare professionals manually assign such codes to track
patient diagnoses and treatments. Automated medical coding can considerably
alleviate this administrative burden. In this paper, we reproduce, compare, and
analyze state-of-the-art automated medical coding machine learning models. We
show that several models underperform due to weak configurations, poorly
sampled train-test splits, and insufficient evaluation. In previous work, the
macro F1 score has been calculated sub-optimally, and our correction doubles
it. We contribute a revised model comparison using stratified sampling and
identical experimental setups, including hyperparameters and decision boundary
tuning. We analyze prediction errors to validate and falsify assumptions of
previous works. The analysis confirms that all models struggle with rare codes,
while long documents only have a negligible impact. Finally, we present the
first comprehensive results on the newly released MIMIC-IV dataset using the
reproduced models. We release our code, model parameters, and new MIMIC-III and
MIMIC-IV training and evaluation pipelines to accommodate fair future
comparisons.
| new_dataset | 0.71403 |
2304.10933 | Romain Menegaux | Romain Menegaux and Emmanuel Jehanno and Margot Selosse and Julien
Mairal | Self-Attention in Colors: Another Take on Encoding Graph Structure in
Transformers | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We introduce a novel self-attention mechanism, which we call CSA (Chromatic
Self-Attention), which extends the notion of attention scores to attention
_filters_, independently modulating the feature channels. We showcase CSA in a
fully-attentional graph Transformer CGT (Chromatic Graph Transformer) which
integrates both graph structural information and edge features, completely
bypassing the need for local message-passing components. Our method flexibly
encodes graph structure through node-node interactions, by enriching the
original edge features with a relative positional encoding scheme. We propose a
new scheme based on random walks that encodes both structural and positional
information, and show how to incorporate higher-order topological information,
such as rings in molecular graphs. Our approach achieves state-of-the-art
results on the ZINC benchmark dataset, while providing a flexible framework for
encoding graph structure and incorporating higher-order topology.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 13:08:53 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Menegaux",
"Romain",
""
],
[
"Jehanno",
"Emmanuel",
""
],
[
"Selosse",
"Margot",
""
],
[
"Mairal",
"Julien",
""
]
] | TITLE: Self-Attention in Colors: Another Take on Encoding Graph Structure in
Transformers
ABSTRACT: We introduce a novel self-attention mechanism, which we call CSA (Chromatic
Self-Attention), which extends the notion of attention scores to attention
_filters_, independently modulating the feature channels. We showcase CSA in a
fully-attentional graph Transformer CGT (Chromatic Graph Transformer) which
integrates both graph structural information and edge features, completely
bypassing the need for local message-passing components. Our method flexibly
encodes graph structure through node-node interactions, by enriching the
original edge features with a relative positional encoding scheme. We propose a
new scheme based on random walks that encodes both structural and positional
information, and show how to incorporate higher-order topological information,
such as rings in molecular graphs. Our approach achieves state-of-the-art
results on the ZINC benchmark dataset, while providing a flexible framework for
encoding graph structure and incorporating higher-order topology.
| no_new_dataset | 0.710007 |
2304.10939 | Marion Neumeier | Marion Neumeier, Andreas Tollk\"uhn, Sebastian Dorn, Michael Botsch,
Wolfgang Utschick | Gradient Derivation for Learnable Parameters in Graph Attention Networks | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work provides a comprehensive derivation of the parameter gradients for
GATv2 [4], a widely used implementation of Graph Attention Networks (GATs).
GATs have proven to be powerful frameworks for processing graph-structured data
and, hence, have been used in a range of applications. However, the achieved
performance by these attempts has been found to be inconsistent across
different datasets and the reasons for this remains an open research question.
As the gradient flow provides valuable insights into the training dynamics of
statistically learning models, this work obtains the gradients for the
trainable model parameters of GATv2. The gradient derivations supplement the
efforts of [2], where potential pitfalls of GATv2 are investigated.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 13:23:38 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Neumeier",
"Marion",
""
],
[
"Tollkühn",
"Andreas",
""
],
[
"Dorn",
"Sebastian",
""
],
[
"Botsch",
"Michael",
""
],
[
"Utschick",
"Wolfgang",
""
]
] | TITLE: Gradient Derivation for Learnable Parameters in Graph Attention Networks
ABSTRACT: This work provides a comprehensive derivation of the parameter gradients for
GATv2 [4], a widely used implementation of Graph Attention Networks (GATs).
GATs have proven to be powerful frameworks for processing graph-structured data
and, hence, have been used in a range of applications. However, the achieved
performance by these attempts has been found to be inconsistent across
different datasets and the reasons for this remains an open research question.
As the gradient flow provides valuable insights into the training dynamics of
statistically learning models, this work obtains the gradients for the
trainable model parameters of GATv2. The gradient derivations supplement the
efforts of [2], where potential pitfalls of GATv2 are investigated.
| no_new_dataset | 0.709466 |
2304.10941 | Zetong Liu | Liu Pingping, Liu Zetong, Lang Yijun, Zhou Qiuzhan, Li Qingliang | Deep Metric Learning Assisted by Intra-variance in A Semi-supervised
View of Learning | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Deep metric learning aims to construct an embedding space where samples of
the same class are close to each other, while samples of different classes are
far away from each other. Most existing deep metric learning methods attempt to
maximize the difference of inter-class features. And semantic related
information is obtained by increasing the distance between samples of different
classes in the embedding space. However, compressing all positive samples
together while creating large margins between different classes unconsciously
destroys the local structure between similar samples. Ignoring the intra-class
variance contained in the local structure between similar samples, the
embedding space obtained from training receives lower generalizability over
unseen classes, which would lead to the network overfitting the training set
and crashing on the test set. To address these considerations, this paper
designs a self-supervised generative assisted ranking framework that provides a
semi-supervised view of intra-class variance learning scheme for typical
supervised deep metric learning. Specifically, this paper performs sample
synthesis with different intensities and diversity for samples satisfying
certain conditions to simulate the complex transformation of intra-class
samples. And an intra-class ranking loss function is designed using the idea of
self-supervised learning to constrain the network to maintain the intra-class
distribution during the training process to capture the subtle intra-class
variance. With this approach, a more realistic embedding space can be obtained
in which global and local structures of samples are well preserved, thus
enhancing the effectiveness of downstream tasks. Extensive experiments on four
benchmarks have shown that this approach surpasses state-of-the-art methods
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 13:30:32 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Pingping",
"Liu",
""
],
[
"Zetong",
"Liu",
""
],
[
"Yijun",
"Lang",
""
],
[
"Qiuzhan",
"Zhou",
""
],
[
"Qingliang",
"Li",
""
]
] | TITLE: Deep Metric Learning Assisted by Intra-variance in A Semi-supervised
View of Learning
ABSTRACT: Deep metric learning aims to construct an embedding space where samples of
the same class are close to each other, while samples of different classes are
far away from each other. Most existing deep metric learning methods attempt to
maximize the difference of inter-class features. And semantic related
information is obtained by increasing the distance between samples of different
classes in the embedding space. However, compressing all positive samples
together while creating large margins between different classes unconsciously
destroys the local structure between similar samples. Ignoring the intra-class
variance contained in the local structure between similar samples, the
embedding space obtained from training receives lower generalizability over
unseen classes, which would lead to the network overfitting the training set
and crashing on the test set. To address these considerations, this paper
designs a self-supervised generative assisted ranking framework that provides a
semi-supervised view of intra-class variance learning scheme for typical
supervised deep metric learning. Specifically, this paper performs sample
synthesis with different intensities and diversity for samples satisfying
certain conditions to simulate the complex transformation of intra-class
samples. And an intra-class ranking loss function is designed using the idea of
self-supervised learning to constrain the network to maintain the intra-class
distribution during the training process to capture the subtle intra-class
variance. With this approach, a more realistic embedding space can be obtained
in which global and local structures of samples are well preserved, thus
enhancing the effectiveness of downstream tasks. Extensive experiments on four
benchmarks have shown that this approach surpasses state-of-the-art methods
| no_new_dataset | 0.710465 |
2304.10961 | Jiajia Mi | Jiajia Mi | An Incomplete Tensor Tucker decomposition based Traffic Speed Prediction
Method | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In intelligent transport systems, it is common and inevitable with missing
data. While complete and valid traffic speed data is of great importance to
intelligent transportation systems. A latent factorization-of-tensors (LFT)
model is one of the most attractive approaches to solve missing traffic data
recovery due to its well-scalability. A LFT model achieves optimization usually
via a stochastic gradient descent (SGD) solver, however, the SGD-based LFT
suffers from slow convergence. To deal with this issue, this work integrates
the unique advantages of the proportional-integral-derivative (PID) controller
into a Tucker decomposition based LFT model. It adopts two-fold ideas: a)
adopting tucker decomposition to build a LFT model for achieving a better
recovery accuracy. b) taking the adjusted instance error based on the PID
control theory into the SGD solver to effectively improve convergence rate. Our
experimental studies on two major city traffic road speed datasets show that
the proposed model achieves significant efficiency gain and highly competitive
prediction accuracy.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 13:59:28 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Mi",
"Jiajia",
""
]
] | TITLE: An Incomplete Tensor Tucker decomposition based Traffic Speed Prediction
Method
ABSTRACT: In intelligent transport systems, it is common and inevitable with missing
data. While complete and valid traffic speed data is of great importance to
intelligent transportation systems. A latent factorization-of-tensors (LFT)
model is one of the most attractive approaches to solve missing traffic data
recovery due to its well-scalability. A LFT model achieves optimization usually
via a stochastic gradient descent (SGD) solver, however, the SGD-based LFT
suffers from slow convergence. To deal with this issue, this work integrates
the unique advantages of the proportional-integral-derivative (PID) controller
into a Tucker decomposition based LFT model. It adopts two-fold ideas: a)
adopting tucker decomposition to build a LFT model for achieving a better
recovery accuracy. b) taking the adjusted instance error based on the PID
control theory into the SGD solver to effectively improve convergence rate. Our
experimental studies on two major city traffic road speed datasets show that
the proposed model achieves significant efficiency gain and highly competitive
prediction accuracy.
| no_new_dataset | 0.709447 |
2304.10973 | David Garcia | Segun Taofeek Aroyehun, Lukas Malik, Hannah Metzler, Nikolas Haimerl,
Anna Di Natale, David Garcia | LEIA: Linguistic Embeddings for the Identification of Affect | null | null | null | null | cs.CL cs.AI cs.CY cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The wealth of text data generated by social media has enabled new kinds of
analysis of emotions with language models. These models are often trained on
small and costly datasets of text annotations produced by readers who guess the
emotions expressed by others in social media posts. This affects the quality of
emotion identification methods due to training data size limitations and noise
in the production of labels used in model development. We present LEIA, a model
for emotion identification in text that has been trained on a dataset of more
than 6 million posts with self-annotated emotion labels for happiness,
affection, sadness, anger, and fear. LEIA is based on a word masking method
that enhances the learning of emotion words during model pre-training. LEIA
achieves macro-F1 values of approximately 73 on three in-domain test datasets,
outperforming other supervised and unsupervised methods in a strong benchmark
that shows that LEIA generalizes across posts, users, and time periods. We
further perform an out-of-domain evaluation on five different datasets of
social media and other sources, showing LEIA's robust performance across media,
data collection methods, and annotation schemes. Our results show that LEIA
generalizes its classification of anger, happiness, and sadness beyond the
domain it was trained on. LEIA can be applied in future research to provide
better identification of emotions in text from the perspective of the writer.
The models produced for this article are publicly available at
https://huggingface.co/LEIA
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 14:17:10 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Aroyehun",
"Segun Taofeek",
""
],
[
"Malik",
"Lukas",
""
],
[
"Metzler",
"Hannah",
""
],
[
"Haimerl",
"Nikolas",
""
],
[
"Di Natale",
"Anna",
""
],
[
"Garcia",
"David",
""
]
] | TITLE: LEIA: Linguistic Embeddings for the Identification of Affect
ABSTRACT: The wealth of text data generated by social media has enabled new kinds of
analysis of emotions with language models. These models are often trained on
small and costly datasets of text annotations produced by readers who guess the
emotions expressed by others in social media posts. This affects the quality of
emotion identification methods due to training data size limitations and noise
in the production of labels used in model development. We present LEIA, a model
for emotion identification in text that has been trained on a dataset of more
than 6 million posts with self-annotated emotion labels for happiness,
affection, sadness, anger, and fear. LEIA is based on a word masking method
that enhances the learning of emotion words during model pre-training. LEIA
achieves macro-F1 values of approximately 73 on three in-domain test datasets,
outperforming other supervised and unsupervised methods in a strong benchmark
that shows that LEIA generalizes across posts, users, and time periods. We
further perform an out-of-domain evaluation on five different datasets of
social media and other sources, showing LEIA's robust performance across media,
data collection methods, and annotation schemes. Our results show that LEIA
generalizes its classification of anger, happiness, and sadness beyond the
domain it was trained on. LEIA can be applied in future research to provide
better identification of emotions in text from the perspective of the writer.
The models produced for this article are publicly available at
https://huggingface.co/LEIA
| new_dataset | 0.721973 |
2304.10978 | Arnaud Delaunoy | Arnaud Delaunoy, Benjamin Kurt Miller, Patrick Forr\'e, Christoph
Weniger, Gilles Louppe | Balancing Simulation-based Inference for Conservative Posteriors | null | null | null | null | stat.ML cs.LG stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Conservative inference is a major concern in simulation-based inference. It
has been shown that commonly used algorithms can produce overconfident
posterior approximations. Balancing has empirically proven to be an effective
way to mitigate this issue. However, its application remains limited to neural
ratio estimation. In this work, we extend balancing to any algorithm that
provides a posterior density. In particular, we introduce a balanced version of
both neural posterior estimation and contrastive neural ratio estimation. We
show empirically that the balanced versions tend to produce conservative
posterior approximations on a wide variety of benchmarks. In addition, we
provide an alternative interpretation of the balancing condition in terms of
the $\chi^2$ divergence.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 14:26:16 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Delaunoy",
"Arnaud",
""
],
[
"Miller",
"Benjamin Kurt",
""
],
[
"Forré",
"Patrick",
""
],
[
"Weniger",
"Christoph",
""
],
[
"Louppe",
"Gilles",
""
]
] | TITLE: Balancing Simulation-based Inference for Conservative Posteriors
ABSTRACT: Conservative inference is a major concern in simulation-based inference. It
has been shown that commonly used algorithms can produce overconfident
posterior approximations. Balancing has empirically proven to be an effective
way to mitigate this issue. However, its application remains limited to neural
ratio estimation. In this work, we extend balancing to any algorithm that
provides a posterior density. In particular, we introduce a balanced version of
both neural posterior estimation and contrastive neural ratio estimation. We
show empirically that the balanced versions tend to produce conservative
posterior approximations on a wide variety of benchmarks. In addition, we
provide an alternative interpretation of the balancing condition in terms of
the $\chi^2$ divergence.
| no_new_dataset | 0.711249 |
2304.10983 | Sergey Titov | Alexander Agroskin, Elena Lyulina, Sergey Titov, Vladimir Kovalenko | Constructing Temporal Networks of OSS Programming Language Ecosystems | Accepted to SANER 2023 | null | null | null | cs.SE cs.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | One of the primary factors that encourage developers to contribute to open
source software (OSS) projects is the collaborative nature of OSS development.
However, the collaborative structure of these communities largely remains
unclear, partly due to the enormous scale of data to be gathered, processed,
and analyzed. In this work, we utilize the World Of Code dataset, which
contains commit activity data for millions of OSS projects, to build
collaboration networks for ten popular programming language ecosystems,
containing in total over 290M commits across over 18M projects. We build a
collaboration graph representation for each language ecosystem, having authors
and projects as nodes, which enables various forms of social network analysis
on the scale of language ecosystems. Moreover, we capture the information on
the ecosystems' evolution by slicing each network into 30 historical snapshots.
Additionally, we calculate multiple collaboration metrics that characterize the
ecosystems' states. We make the resulting dataset publicly available, including
the constructed graphs and the pipeline enabling the analysis of more
ecosystems.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 14:30:30 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Agroskin",
"Alexander",
""
],
[
"Lyulina",
"Elena",
""
],
[
"Titov",
"Sergey",
""
],
[
"Kovalenko",
"Vladimir",
""
]
] | TITLE: Constructing Temporal Networks of OSS Programming Language Ecosystems
ABSTRACT: One of the primary factors that encourage developers to contribute to open
source software (OSS) projects is the collaborative nature of OSS development.
However, the collaborative structure of these communities largely remains
unclear, partly due to the enormous scale of data to be gathered, processed,
and analyzed. In this work, we utilize the World Of Code dataset, which
contains commit activity data for millions of OSS projects, to build
collaboration networks for ten popular programming language ecosystems,
containing in total over 290M commits across over 18M projects. We build a
collaboration graph representation for each language ecosystem, having authors
and projects as nodes, which enables various forms of social network analysis
on the scale of language ecosystems. Moreover, we capture the information on
the ecosystems' evolution by slicing each network into 30 historical snapshots.
Additionally, we calculate multiple collaboration metrics that characterize the
ecosystems' states. We make the resulting dataset publicly available, including
the constructed graphs and the pipeline enabling the analysis of more
ecosystems.
| new_dataset | 0.714864 |
2304.10987 | Jianheng Liu | Jianheng Liu, Xuanfu Li, Yueqian Liu, Haoyao Chen | RGB-D Inertial Odometry for a Resource-Restricted Robot in Dynamic
Environments | null | IEEE Robotics and Automation Letters ( Volume: 7, Issue: 4,
October 2022) | 10.1109/LRA.2022.3191193 | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Current simultaneous localization and mapping (SLAM) algorithms perform well
in static environments but easily fail in dynamic environments. Recent works
introduce deep learning-based semantic information to SLAM systems to reduce
the influence of dynamic objects. However, it is still challenging to apply a
robust localization in dynamic environments for resource-restricted robots.
This paper proposes a real-time RGB-D inertial odometry system for
resource-restricted robots in dynamic environments named Dynamic-VINS. Three
main threads run in parallel: object detection, feature tracking, and state
optimization. The proposed Dynamic-VINS combines object detection and depth
information for dynamic feature recognition and achieves performance comparable
to semantic segmentation. Dynamic-VINS adopts grid-based feature detection and
proposes a fast and efficient method to extract high-quality FAST feature
points. IMU is applied to predict motion for feature tracking and moving
consistency check. The proposed method is evaluated on both public datasets and
real-world applications and shows competitive localization accuracy and
robustness in dynamic environments. Yet, to the best of our knowledge, it is
the best-performance real-time RGB-D inertial odometry for resource-restricted
platforms in dynamic environments for now. The proposed system is open source
at: https://github.com/HITSZ-NRSL/Dynamic-VINS.git
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 14:37:11 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Liu",
"Jianheng",
""
],
[
"Li",
"Xuanfu",
""
],
[
"Liu",
"Yueqian",
""
],
[
"Chen",
"Haoyao",
""
]
] | TITLE: RGB-D Inertial Odometry for a Resource-Restricted Robot in Dynamic
Environments
ABSTRACT: Current simultaneous localization and mapping (SLAM) algorithms perform well
in static environments but easily fail in dynamic environments. Recent works
introduce deep learning-based semantic information to SLAM systems to reduce
the influence of dynamic objects. However, it is still challenging to apply a
robust localization in dynamic environments for resource-restricted robots.
This paper proposes a real-time RGB-D inertial odometry system for
resource-restricted robots in dynamic environments named Dynamic-VINS. Three
main threads run in parallel: object detection, feature tracking, and state
optimization. The proposed Dynamic-VINS combines object detection and depth
information for dynamic feature recognition and achieves performance comparable
to semantic segmentation. Dynamic-VINS adopts grid-based feature detection and
proposes a fast and efficient method to extract high-quality FAST feature
points. IMU is applied to predict motion for feature tracking and moving
consistency check. The proposed method is evaluated on both public datasets and
real-world applications and shows competitive localization accuracy and
robustness in dynamic environments. Yet, to the best of our knowledge, it is
the best-performance real-time RGB-D inertial odometry for resource-restricted
platforms in dynamic environments for now. The proposed system is open source
at: https://github.com/HITSZ-NRSL/Dynamic-VINS.git
| no_new_dataset | 0.711281 |
2304.10994 | Laurent Lam | Laurent Lam, Pirashanth Ratnamogan, Jo\"el Tang, William Vanhuffel and
Fabien Caspani | Information Extraction from Documents: Question Answering vs Token
Classification in real-world setups | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Research in Document Intelligence and especially in Document Key Information
Extraction (DocKIE) has been mainly solved as Token Classification problem.
Recent breakthroughs in both natural language processing (NLP) and computer
vision helped building document-focused pre-training methods, leveraging a
multimodal understanding of the document text, layout and image modalities.
However, these breakthroughs also led to the emergence of a new DocKIE subtask
of extractive document Question Answering (DocQA), as part of the Machine
Reading Comprehension (MRC) research field. In this work, we compare the
Question Answering approach with the classical token classification approach
for document key information extraction. We designed experiments to benchmark
five different experimental setups : raw performances, robustness to noisy
environment, capacity to extract long entities, fine-tuning speed on Few-Shot
Learning and finally Zero-Shot Learning. Our research showed that when dealing
with clean and relatively short entities, it is still best to use token
classification-based approach, while the QA approach could be a good
alternative for noisy environment or long entities use-cases.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 14:43:42 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Lam",
"Laurent",
""
],
[
"Ratnamogan",
"Pirashanth",
""
],
[
"Tang",
"Joël",
""
],
[
"Vanhuffel",
"William",
""
],
[
"Caspani",
"Fabien",
""
]
] | TITLE: Information Extraction from Documents: Question Answering vs Token
Classification in real-world setups
ABSTRACT: Research in Document Intelligence and especially in Document Key Information
Extraction (DocKIE) has been mainly solved as Token Classification problem.
Recent breakthroughs in both natural language processing (NLP) and computer
vision helped building document-focused pre-training methods, leveraging a
multimodal understanding of the document text, layout and image modalities.
However, these breakthroughs also led to the emergence of a new DocKIE subtask
of extractive document Question Answering (DocQA), as part of the Machine
Reading Comprehension (MRC) research field. In this work, we compare the
Question Answering approach with the classical token classification approach
for document key information extraction. We designed experiments to benchmark
five different experimental setups : raw performances, robustness to noisy
environment, capacity to extract long entities, fine-tuning speed on Few-Shot
Learning and finally Zero-Shot Learning. Our research showed that when dealing
with clean and relatively short entities, it is still best to use token
classification-based approach, while the QA approach could be a good
alternative for noisy environment or long entities use-cases.
| no_new_dataset | 0.709453 |
2304.11020 | Jake Stuchbury-Wass | Jake Stuchbury-Wass, Erika Bondareva, Kayla-Jade Butkow, Sanja
Scepanovic, Zoran Radivojevic and Cecilia Mascolo | Heart Rate Extraction from Abdominal Audio Signals | ICASSP 2023 | null | null | null | eess.AS cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Abdominal sounds (ABS) have been traditionally used for assessing
gastrointestinal (GI) disorders. However, the assessment requires a trained
medical professional to perform multiple abdominal auscultation sessions, which
is resource-intense and may fail to provide an accurate picture of patients'
continuous GI wellbeing. This has generated a technological interest in
developing wearables for continuous capture of ABS, which enables a fuller
picture of patient's GI status to be obtained at reduced cost. This paper seeks
to evaluate the feasibility of extracting heart rate (HR) from such ABS
monitoring devices. The collection of HR directly from these devices would
enable gathering vital signs alongside GI data without the need for additional
wearable devices, providing further cost benefits and improving general
usability. We utilised a dataset containing 104 hours of ABS audio, collected
from the abdomen using an e-stethoscope, and electrocardiogram as ground truth.
Our evaluation shows for the first time that we can successfully extract HR
from audio collected from a wearable on the abdomen. As heart sounds collected
from the abdomen suffer from significant noise from GI and respiratory tracts,
we leverage wavelet denoising for improved heart beat detection. The mean
absolute error of the algorithm for average HR is 3.4 BPM with mean directional
error of -1.2 BPM over the whole dataset. A comparison to
photoplethysmography-based wearable HR sensors shows that our approach exhibits
comparable accuracy to consumer wrist-worn wearables for average and
instantaneous heart rate.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 15:08:15 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Stuchbury-Wass",
"Jake",
""
],
[
"Bondareva",
"Erika",
""
],
[
"Butkow",
"Kayla-Jade",
""
],
[
"Scepanovic",
"Sanja",
""
],
[
"Radivojevic",
"Zoran",
""
],
[
"Mascolo",
"Cecilia",
""
]
] | TITLE: Heart Rate Extraction from Abdominal Audio Signals
ABSTRACT: Abdominal sounds (ABS) have been traditionally used for assessing
gastrointestinal (GI) disorders. However, the assessment requires a trained
medical professional to perform multiple abdominal auscultation sessions, which
is resource-intense and may fail to provide an accurate picture of patients'
continuous GI wellbeing. This has generated a technological interest in
developing wearables for continuous capture of ABS, which enables a fuller
picture of patient's GI status to be obtained at reduced cost. This paper seeks
to evaluate the feasibility of extracting heart rate (HR) from such ABS
monitoring devices. The collection of HR directly from these devices would
enable gathering vital signs alongside GI data without the need for additional
wearable devices, providing further cost benefits and improving general
usability. We utilised a dataset containing 104 hours of ABS audio, collected
from the abdomen using an e-stethoscope, and electrocardiogram as ground truth.
Our evaluation shows for the first time that we can successfully extract HR
from audio collected from a wearable on the abdomen. As heart sounds collected
from the abdomen suffer from significant noise from GI and respiratory tracts,
we leverage wavelet denoising for improved heart beat detection. The mean
absolute error of the algorithm for average HR is 3.4 BPM with mean directional
error of -1.2 BPM over the whole dataset. A comparison to
photoplethysmography-based wearable HR sensors shows that our approach exhibits
comparable accuracy to consumer wrist-worn wearables for average and
instantaneous heart rate.
| new_dataset | 0.721985 |
2304.11040 | Gazi Zahirul Islam | Mehrab Hosain, Most. Yeasmin Arafat, Gazi Zahirul Islam, Jia Uddin,
Md. Mobarak Hossain, Fatema Alam | Emotional Expression Detection in Spoken Language Employing Machine
Learning Algorithms | Journal Pre-print (15 Pages, 9 Figures, 3 Tables) | null | null | null | cs.SD cs.CL cs.LG eess.AS | http://creativecommons.org/licenses/by/4.0/ | There are a variety of features of the human voice that can be classified as
pitch, timbre, loudness, and vocal tone. It is observed in numerous incidents
that human expresses their feelings using different vocal qualities when they
are speaking. The primary objective of this research is to recognize different
emotions of human beings such as anger, sadness, fear, neutrality, disgust,
pleasant surprise, and happiness by using several MATLAB functions namely,
spectral descriptors, periodicity, and harmonicity. To accomplish the work, we
analyze the CREMA-D (Crowd-sourced Emotional Multimodal Actors Data) & TESS
(Toronto Emotional Speech Set) datasets of human speech. The audio file
contains data that have various characteristics (e.g., noisy, speedy, slow)
thereby the efficiency of the ML (Machine Learning) models increases
significantly. The EMD (Empirical Mode Decomposition) is utilized for the
process of signal decomposition. Then, the features are extracted through the
use of several techniques such as the MFCC, GTCC, spectral centroid, roll-off
point, entropy, spread, flux, harmonic ratio, energy, skewness, flatness, and
audio delta. The data is trained using some renowned ML models namely, Support
Vector Machine, Neural Network, Ensemble, and KNN. The algorithms show an
accuracy of 67.7%, 63.3%, 61.6%, and 59.0% respectively for the test data and
77.7%, 76.1%, 99.1%, and 61.2% for the training data. We have conducted
experiments using Matlab and the result shows that our model is very prominent
and flexible than existing similar works.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 17:57:08 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Hosain",
"Mehrab",
""
],
[
"Arafat",
"Most. Yeasmin",
""
],
[
"Islam",
"Gazi Zahirul",
""
],
[
"Uddin",
"Jia",
""
],
[
"Hossain",
"Md. Mobarak",
""
],
[
"Alam",
"Fatema",
""
]
] | TITLE: Emotional Expression Detection in Spoken Language Employing Machine
Learning Algorithms
ABSTRACT: There are a variety of features of the human voice that can be classified as
pitch, timbre, loudness, and vocal tone. It is observed in numerous incidents
that human expresses their feelings using different vocal qualities when they
are speaking. The primary objective of this research is to recognize different
emotions of human beings such as anger, sadness, fear, neutrality, disgust,
pleasant surprise, and happiness by using several MATLAB functions namely,
spectral descriptors, periodicity, and harmonicity. To accomplish the work, we
analyze the CREMA-D (Crowd-sourced Emotional Multimodal Actors Data) & TESS
(Toronto Emotional Speech Set) datasets of human speech. The audio file
contains data that have various characteristics (e.g., noisy, speedy, slow)
thereby the efficiency of the ML (Machine Learning) models increases
significantly. The EMD (Empirical Mode Decomposition) is utilized for the
process of signal decomposition. Then, the features are extracted through the
use of several techniques such as the MFCC, GTCC, spectral centroid, roll-off
point, entropy, spread, flux, harmonic ratio, energy, skewness, flatness, and
audio delta. The data is trained using some renowned ML models namely, Support
Vector Machine, Neural Network, Ensemble, and KNN. The algorithms show an
accuracy of 67.7%, 63.3%, 61.6%, and 59.0% respectively for the test data and
77.7%, 76.1%, 99.1%, and 61.2% for the training data. We have conducted
experiments using Matlab and the result shows that our model is very prominent
and flexible than existing similar works.
| no_new_dataset | 0.709025 |
2304.11045 | Pawan Kumar | Istasis Mishra, Arpan Dasgupta, Pratik Jawanpuria, Bamdev Mishra, and
Pawan Kumar | Light-weight Deep Extreme Multilabel Classification | 9 pages, 2 figures, 5 tables | null | null | null | cs.LG cs.AI cs.IR | http://creativecommons.org/licenses/by/4.0/ | Extreme multi-label (XML) classification refers to the task of supervised
multi-label learning that involves a large number of labels. Hence, scalability
of the classifier with increasing label dimension is an important
consideration. In this paper, we develop a method called LightDXML which
modifies the recently developed deep learning based XML framework by using
label embeddings instead of feature embedding for negative sampling and
iterating cyclically through three major phases: (1) proxy training of label
embeddings (2) shortlisting of labels for negative sampling and (3) final
classifier training using the negative samples. Consequently, LightDXML also
removes the requirement of a re-ranker module, thereby, leading to further
savings on time and memory requirements. The proposed method achieves the best
of both worlds: while the training time, model size and prediction times are on
par or better compared to the tree-based methods, it attains much better
prediction accuracy that is on par with the deep learning based methods.
Moreover, the proposed approach achieves the best tail-label prediction
accuracy over most state-of-the-art XML methods on some of the large
datasets\footnote{accepted in IJCNN 2023, partial funding from MAPG grant and
IIIT Seed grant at IIIT, Hyderabad, India. Code:
\url{https://github.com/misterpawan/LightDXML}
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 09:06:10 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Mishra",
"Istasis",
""
],
[
"Dasgupta",
"Arpan",
""
],
[
"Jawanpuria",
"Pratik",
""
],
[
"Mishra",
"Bamdev",
""
],
[
"Kumar",
"Pawan",
""
]
] | TITLE: Light-weight Deep Extreme Multilabel Classification
ABSTRACT: Extreme multi-label (XML) classification refers to the task of supervised
multi-label learning that involves a large number of labels. Hence, scalability
of the classifier with increasing label dimension is an important
consideration. In this paper, we develop a method called LightDXML which
modifies the recently developed deep learning based XML framework by using
label embeddings instead of feature embedding for negative sampling and
iterating cyclically through three major phases: (1) proxy training of label
embeddings (2) shortlisting of labels for negative sampling and (3) final
classifier training using the negative samples. Consequently, LightDXML also
removes the requirement of a re-ranker module, thereby, leading to further
savings on time and memory requirements. The proposed method achieves the best
of both worlds: while the training time, model size and prediction times are on
par or better compared to the tree-based methods, it attains much better
prediction accuracy that is on par with the deep learning based methods.
Moreover, the proposed approach achieves the best tail-label prediction
accuracy over most state-of-the-art XML methods on some of the large
datasets\footnote{accepted in IJCNN 2023, partial funding from MAPG grant and
IIIT Seed grant at IIIT, Hyderabad, India. Code:
\url{https://github.com/misterpawan/LightDXML}
| no_new_dataset | 0.71103 |
2304.11058 | Ankan Mullick | Ankan Mullick | Novel Intent Detection and Active Learning Based Classification (Student
Abstract) | AAAI 2023 Student Abstract | null | null | null | cs.CL cs.IR | http://creativecommons.org/licenses/by/4.0/ | Novel intent class detection is an important problem in real world scenario
for conversational agents for continuous interaction. Several research works
have been done to detect novel intents in a mono-lingual (primarily English)
texts and images. But, current systems lack an end-to-end universal framework
to detect novel intents across various different languages with less human
annotation effort for mis-classified and system rejected samples. This paper
proposes NIDAL (Novel Intent Detection and Active Learning based
classification), a semi-supervised framework to detect novel intents while
reducing human annotation cost. Empirical results on various benchmark datasets
demonstrate that this system outperforms the baseline methods by more than 10%
margin for accuracy and macro-F1. The system achieves this while maintaining
overall annotation cost to be just ~6-10% of the unlabeled data available to
the system.
| [
{
"version": "v1",
"created": "Wed, 22 Feb 2023 13:04:17 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Mullick",
"Ankan",
""
]
] | TITLE: Novel Intent Detection and Active Learning Based Classification (Student
Abstract)
ABSTRACT: Novel intent class detection is an important problem in real world scenario
for conversational agents for continuous interaction. Several research works
have been done to detect novel intents in a mono-lingual (primarily English)
texts and images. But, current systems lack an end-to-end universal framework
to detect novel intents across various different languages with less human
annotation effort for mis-classified and system rejected samples. This paper
proposes NIDAL (Novel Intent Detection and Active Learning based
classification), a semi-supervised framework to detect novel intents while
reducing human annotation cost. Empirical results on various benchmark datasets
demonstrate that this system outperforms the baseline methods by more than 10%
margin for accuracy and macro-F1. The system achieves this while maintaining
overall annotation cost to be just ~6-10% of the unlabeled data available to
the system.
| no_new_dataset | 0.710867 |
2304.11061 | Mingjun Zhao | Mingjun Zhao, Mengzhen Wang, Yinglong Ma, Di Niu and Haijiang Wu | CEIL: A General Classification-Enhanced Iterative Learning Framework for
Text Clustering | The Web Conference 2023 | null | 10.1145/3543507.3583457 | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Text clustering, as one of the most fundamental challenges in unsupervised
learning, aims at grouping semantically similar text segments without relying
on human annotations. With the rapid development of deep learning, deep
clustering has achieved significant advantages over traditional clustering
methods. Despite the effectiveness, most existing deep text clustering methods
rely heavily on representations pre-trained in general domains, which may not
be the most suitable solution for clustering in specific target domains. To
address this issue, we propose CEIL, a novel Classification-Enhanced Iterative
Learning framework for short text clustering, which aims at generally promoting
the clustering performance by introducing a classification objective to
iteratively improve feature representations. In each iteration, we first adopt
a language model to retrieve the initial text representations, from which the
clustering results are collected using our proposed Category Disentangled
Contrastive Clustering (CDCC) algorithm. After strict data filtering and
aggregation processes, samples with clean category labels are retrieved, which
serve as supervision information to update the language model with the
classification objective via a prompt learning approach. Finally, the updated
language model with improved representation ability is used to enhance
clustering in the next iteration. Extensive experiments demonstrate that the
CEIL framework significantly improves the clustering performance over
iterations, and is generally effective on various clustering algorithms.
Moreover, by incorporating CEIL on CDCC, we achieve the state-of-the-art
clustering performance on a wide range of short text clustering benchmarks
outperforming other strong baseline methods.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 14:04:31 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Zhao",
"Mingjun",
""
],
[
"Wang",
"Mengzhen",
""
],
[
"Ma",
"Yinglong",
""
],
[
"Niu",
"Di",
""
],
[
"Wu",
"Haijiang",
""
]
] | TITLE: CEIL: A General Classification-Enhanced Iterative Learning Framework for
Text Clustering
ABSTRACT: Text clustering, as one of the most fundamental challenges in unsupervised
learning, aims at grouping semantically similar text segments without relying
on human annotations. With the rapid development of deep learning, deep
clustering has achieved significant advantages over traditional clustering
methods. Despite the effectiveness, most existing deep text clustering methods
rely heavily on representations pre-trained in general domains, which may not
be the most suitable solution for clustering in specific target domains. To
address this issue, we propose CEIL, a novel Classification-Enhanced Iterative
Learning framework for short text clustering, which aims at generally promoting
the clustering performance by introducing a classification objective to
iteratively improve feature representations. In each iteration, we first adopt
a language model to retrieve the initial text representations, from which the
clustering results are collected using our proposed Category Disentangled
Contrastive Clustering (CDCC) algorithm. After strict data filtering and
aggregation processes, samples with clean category labels are retrieved, which
serve as supervision information to update the language model with the
classification objective via a prompt learning approach. Finally, the updated
language model with improved representation ability is used to enhance
clustering in the next iteration. Extensive experiments demonstrate that the
CEIL framework significantly improves the clustering performance over
iterations, and is generally effective on various clustering algorithms.
Moreover, by incorporating CEIL on CDCC, we achieve the state-of-the-art
clustering performance on a wide range of short text clustering benchmarks
outperforming other strong baseline methods.
| no_new_dataset | 0.710396 |
2304.11063 | Lina Mezghani | Lina Mezghani and Piotr Bojanowski and Karteek Alahari and Sainbayar
Sukhbaatar | Think Before You Act: Unified Policy for Interleaving Language Reasoning
with Actions | null | Reincarnating Reinforcement Learning Workshop at ICLR 2023 | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The success of transformer models trained with a language modeling objective
brings a promising opportunity to the reinforcement learning framework.
Decision Transformer is a step towards this direction, showing how to train
transformers with a similar next-step prediction objective on offline data.
Another important development in this area is the recent emergence of
large-scale datasets collected from the internet, such as the ones composed of
tutorial videos with captions where people talk about what they are doing. To
take advantage of this language component, we propose a novel method for
unifying language reasoning with actions in a single policy. Specifically, we
augment a transformer policy with word outputs, so it can generate textual
captions interleaved with actions. When tested on the most challenging task in
BabyAI, with captions describing next subgoals, our reasoning policy
consistently outperforms the caption-free baseline.
| [
{
"version": "v1",
"created": "Tue, 18 Apr 2023 16:12:38 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Mezghani",
"Lina",
""
],
[
"Bojanowski",
"Piotr",
""
],
[
"Alahari",
"Karteek",
""
],
[
"Sukhbaatar",
"Sainbayar",
""
]
] | TITLE: Think Before You Act: Unified Policy for Interleaving Language Reasoning
with Actions
ABSTRACT: The success of transformer models trained with a language modeling objective
brings a promising opportunity to the reinforcement learning framework.
Decision Transformer is a step towards this direction, showing how to train
transformers with a similar next-step prediction objective on offline data.
Another important development in this area is the recent emergence of
large-scale datasets collected from the internet, such as the ones composed of
tutorial videos with captions where people talk about what they are doing. To
take advantage of this language component, we propose a novel method for
unifying language reasoning with actions in a single policy. Specifically, we
augment a transformer policy with word outputs, so it can generate textual
captions interleaved with actions. When tested on the most challenging task in
BabyAI, with captions describing next subgoals, our reasoning policy
consistently outperforms the caption-free baseline.
| no_new_dataset | 0.704109 |
2304.11077 | Vitaly Shalumov | Vitaly Shalumov and Harel Haskey | HeRo: RoBERTa and Longformer Hebrew Language Models | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we fill in an existing gap in resources available to the
Hebrew NLP community by providing it with the largest so far pre-train dataset
HeDC4, a state-of-the-art pre-trained language model HeRo for standard length
inputs and an efficient transformer LongHeRo for long input sequences. The HeRo
model was evaluated on the sentiment analysis, the named entity recognition,
and the question answering tasks while the LongHeRo model was evaluated on the
document classification task with a dataset composed of long documents. Both
HeRo and LongHeRo presented state-of-the-art performance. The dataset and model
checkpoints used in this work are publicly available.
| [
{
"version": "v1",
"created": "Tue, 18 Apr 2023 05:56:32 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Shalumov",
"Vitaly",
""
],
[
"Haskey",
"Harel",
""
]
] | TITLE: HeRo: RoBERTa and Longformer Hebrew Language Models
ABSTRACT: In this paper, we fill in an existing gap in resources available to the
Hebrew NLP community by providing it with the largest so far pre-train dataset
HeDC4, a state-of-the-art pre-trained language model HeRo for standard length
inputs and an efficient transformer LongHeRo for long input sequences. The HeRo
model was evaluated on the sentiment analysis, the named entity recognition,
and the question answering tasks while the LongHeRo model was evaluated on the
document classification task with a dataset composed of long documents. Both
HeRo and LongHeRo presented state-of-the-art performance. The dataset and model
checkpoints used in this work are publicly available.
| new_dataset | 0.695958 |
2304.11091 | Qasim Ahmed | Fuhu Che, Qasim Zeeshan Ahmed, Jaron Fontaine, Ben Van Herbruggen,
Adnan Shahid, Eli De Poorter, and Pavlos I. Lazaridis | Feature-Based Generalized Gaussian Distribution Method for NLoS
Detection in Ultra-Wideband (UWB) Indoor Positioning System | null | IEEE Sensors Journal, vol. 22, no. 19, pp. 18726-18739, 1 Oct.1,
2022, | 10.1109/JSEN.2022.3198680. | null | eess.SP cs.LG | http://creativecommons.org/licenses/by/4.0/ | Non-Line-of-Sight (NLoS) propagation condition is a crucial factor affecting
the precision of the localization in the Ultra-Wideband (UWB) Indoor
Positioning System (IPS). Numerous supervised Machine Learning (ML) approaches
have been applied for NLoS identification to improve the accuracy of the IPS.
However, it is difficult for existing ML approaches to maintain a high
classification accuracy when the database contains a small number of NLoS
signals and a large number of Line-of-Sight (LoS) signals. The inaccurate
localization of the target node caused by this small number of NLoS signals can
still be problematic. To solve this issue, we propose feature-based Gaussian
Distribution (GD) and Generalized Gaussian Distribution (GGD) NLoS detection
algorithms. By employing our detection algorithm for the imbalanced dataset, a
classification accuracy of $96.7\%$ and $98.0\%$ can be achieved. We also
compared the proposed algorithm with the existing cutting-edge such as
Support-Vector-Machine (SVM), Decision Tree (DT), Naive Bayes (NB), and Neural
Network (NN), which can achieve an accuracy of $92.6\%$, $92.8\%$, $93.2\%$,
and $95.5\%$, respectively. The results demonstrate that the GGD algorithm can
achieve high classification accuracy with the imbalanced dataset. Finally, the
proposed algorithm can also achieve a higher classification accuracy for
different ratios of LoS and NLoS signals which proves the robustness and
effectiveness of the proposed method.
| [
{
"version": "v1",
"created": "Fri, 14 Apr 2023 11:51:12 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Che",
"Fuhu",
""
],
[
"Ahmed",
"Qasim Zeeshan",
""
],
[
"Fontaine",
"Jaron",
""
],
[
"Van Herbruggen",
"Ben",
""
],
[
"Shahid",
"Adnan",
""
],
[
"De Poorter",
"Eli",
""
],
[
"Lazaridis",
"Pavlos I.",
""
]
] | TITLE: Feature-Based Generalized Gaussian Distribution Method for NLoS
Detection in Ultra-Wideband (UWB) Indoor Positioning System
ABSTRACT: Non-Line-of-Sight (NLoS) propagation condition is a crucial factor affecting
the precision of the localization in the Ultra-Wideband (UWB) Indoor
Positioning System (IPS). Numerous supervised Machine Learning (ML) approaches
have been applied for NLoS identification to improve the accuracy of the IPS.
However, it is difficult for existing ML approaches to maintain a high
classification accuracy when the database contains a small number of NLoS
signals and a large number of Line-of-Sight (LoS) signals. The inaccurate
localization of the target node caused by this small number of NLoS signals can
still be problematic. To solve this issue, we propose feature-based Gaussian
Distribution (GD) and Generalized Gaussian Distribution (GGD) NLoS detection
algorithms. By employing our detection algorithm for the imbalanced dataset, a
classification accuracy of $96.7\%$ and $98.0\%$ can be achieved. We also
compared the proposed algorithm with the existing cutting-edge such as
Support-Vector-Machine (SVM), Decision Tree (DT), Naive Bayes (NB), and Neural
Network (NN), which can achieve an accuracy of $92.6\%$, $92.8\%$, $93.2\%$,
and $95.5\%$, respectively. The results demonstrate that the GGD algorithm can
achieve high classification accuracy with the imbalanced dataset. Finally, the
proposed algorithm can also achieve a higher classification accuracy for
different ratios of LoS and NLoS signals which proves the robustness and
effectiveness of the proposed method.
| no_new_dataset | 0.711588 |
2304.11093 | Meidai Xuanyuan | Meidai Xuanyuan, Yuwang Wang, Honglei Guo, Xiao Ma, Yuchen Guo, Tao
Yu, Qionghai Dai | Hi Sheldon! Creating Deep Personalized Characters from TV Shows | null | null | null | null | cs.CL cs.AI cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Imagine an interesting multimodal interactive scenario that you can see,
hear, and chat with an AI-generated digital character, who is capable of
behaving like Sheldon from The Big Bang Theory, as a DEEP copy from appearance
to personality. Towards this fantastic multimodal chatting scenario, we propose
a novel task, named Deep Personalized Character Creation (DPCC): creating
multimodal chat personalized characters from multimodal data such as TV shows.
Specifically, given a single- or multi-modality input (text, audio, video), the
goal of DPCC is to generate a multi-modality (text, audio, video) response,
which should be well-matched the personality of a specific character such as
Sheldon, and of high quality as well. To support this novel task, we further
collect a character centric multimodal dialogue dataset, named Deep
Personalized Character Dataset (DPCD), from TV shows. DPCD contains
character-specific multimodal dialogue data of ~10k utterances and ~6 hours of
audio/video per character, which is around 10 times larger compared to existing
related datasets.On DPCD, we present a baseline method for the DPCC task and
create 5 Deep personalized digital Characters (DeepCharacters) from Big Bang TV
Shows. We conduct both subjective and objective experiments to evaluate the
multimodal response from DeepCharacters in terms of characterization and
quality. The results demonstrates that, on our collected DPCD dataset, the
proposed baseline can create personalized digital characters for generating
multimodal response.Our collected DPCD dataset, the code of data collection and
our baseline will be published soon.
| [
{
"version": "v1",
"created": "Sun, 9 Apr 2023 00:39:43 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Xuanyuan",
"Meidai",
""
],
[
"Wang",
"Yuwang",
""
],
[
"Guo",
"Honglei",
""
],
[
"Ma",
"Xiao",
""
],
[
"Guo",
"Yuchen",
""
],
[
"Yu",
"Tao",
""
],
[
"Dai",
"Qionghai",
""
]
] | TITLE: Hi Sheldon! Creating Deep Personalized Characters from TV Shows
ABSTRACT: Imagine an interesting multimodal interactive scenario that you can see,
hear, and chat with an AI-generated digital character, who is capable of
behaving like Sheldon from The Big Bang Theory, as a DEEP copy from appearance
to personality. Towards this fantastic multimodal chatting scenario, we propose
a novel task, named Deep Personalized Character Creation (DPCC): creating
multimodal chat personalized characters from multimodal data such as TV shows.
Specifically, given a single- or multi-modality input (text, audio, video), the
goal of DPCC is to generate a multi-modality (text, audio, video) response,
which should be well-matched the personality of a specific character such as
Sheldon, and of high quality as well. To support this novel task, we further
collect a character centric multimodal dialogue dataset, named Deep
Personalized Character Dataset (DPCD), from TV shows. DPCD contains
character-specific multimodal dialogue data of ~10k utterances and ~6 hours of
audio/video per character, which is around 10 times larger compared to existing
related datasets.On DPCD, we present a baseline method for the DPCC task and
create 5 Deep personalized digital Characters (DeepCharacters) from Big Bang TV
Shows. We conduct both subjective and objective experiments to evaluate the
multimodal response from DeepCharacters in terms of characterization and
quality. The results demonstrates that, on our collected DPCD dataset, the
proposed baseline can create personalized digital characters for generating
multimodal response.Our collected DPCD dataset, the code of data collection and
our baseline will be published soon.
| new_dataset | 0.717061 |
2304.11094 | Vithya Yogarajan | Vithya Yogarajan, Gillian Dobbie, Henry Gouk | Effectiveness of Debiasing Techniques: An Indigenous Qualitative
Analysis | accepted with invite to present | ICLR TinyPaper 2023 | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | An indigenous perspective on the effectiveness of debiasing techniques for
pre-trained language models (PLMs) is presented in this paper. The current
techniques used to measure and debias PLMs are skewed towards the US racial
biases and rely on pre-defined bias attributes (e.g. "black" vs "white"). Some
require large datasets and further pre-training. Such techniques are not
designed to capture the underrepresented indigenous populations in other
countries, such as M\=aori in New Zealand. Local knowledge and understanding
must be incorporated to ensure unbiased algorithms, especially when addressing
a resource-restricted society.
| [
{
"version": "v1",
"created": "Mon, 17 Apr 2023 22:25:25 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Yogarajan",
"Vithya",
""
],
[
"Dobbie",
"Gillian",
""
],
[
"Gouk",
"Henry",
""
]
] | TITLE: Effectiveness of Debiasing Techniques: An Indigenous Qualitative
Analysis
ABSTRACT: An indigenous perspective on the effectiveness of debiasing techniques for
pre-trained language models (PLMs) is presented in this paper. The current
techniques used to measure and debias PLMs are skewed towards the US racial
biases and rely on pre-defined bias attributes (e.g. "black" vs "white"). Some
require large datasets and further pre-training. Such techniques are not
designed to capture the underrepresented indigenous populations in other
countries, such as M\=aori in New Zealand. Local knowledge and understanding
must be incorporated to ensure unbiased algorithms, especially when addressing
a resource-restricted society.
| no_new_dataset | 0.711844 |
2304.11101 | Clemens Heistracher | Viktorija Pruckovskaja, Axel Weissenfeld, Clemens Heistracher, Anita
Graser, Julia Kafka, Peter Leputsch, Daniel Schall, Jana Kemnitz | Federated Learning for Predictive Maintenance and Quality Inspection in
Industrial Applications | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data-driven machine learning is playing a crucial role in the advancements of
Industry 4.0, specifically in enhancing predictive maintenance and quality
inspection. Federated learning (FL) enables multiple participants to develop a
machine learning model without compromising the privacy and confidentiality of
their data. In this paper, we evaluate the performance of different FL
aggregation methods and compare them to central and local training approaches.
Our study is based on four datasets with varying data distributions. The
results indicate that the performance of FL is highly dependent on the data and
its distribution among clients. In some scenarios, FL can be an effective
alternative to traditional central or local training methods. Additionally, we
introduce a new federated learning dataset from a real-world quality inspection
setting.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 16:11:09 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Pruckovskaja",
"Viktorija",
""
],
[
"Weissenfeld",
"Axel",
""
],
[
"Heistracher",
"Clemens",
""
],
[
"Graser",
"Anita",
""
],
[
"Kafka",
"Julia",
""
],
[
"Leputsch",
"Peter",
""
],
[
"Schall",
"Daniel",
""
],
[
"Kemnitz",
"Jana",
""
]
] | TITLE: Federated Learning for Predictive Maintenance and Quality Inspection in
Industrial Applications
ABSTRACT: Data-driven machine learning is playing a crucial role in the advancements of
Industry 4.0, specifically in enhancing predictive maintenance and quality
inspection. Federated learning (FL) enables multiple participants to develop a
machine learning model without compromising the privacy and confidentiality of
their data. In this paper, we evaluate the performance of different FL
aggregation methods and compare them to central and local training approaches.
Our study is based on four datasets with varying data distributions. The
results indicate that the performance of FL is highly dependent on the data and
its distribution among clients. In some scenarios, FL can be an effective
alternative to traditional central or local training methods. Additionally, we
introduce a new federated learning dataset from a real-world quality inspection
setting.
| new_dataset | 0.712426 |
2304.11117 | Samir Sadok | Samir Sadok, Simon Leglaive, Renaud S\'eguier | A vector quantized masked autoencoder for speech emotion recognition | https://samsad35.github.io/VQ-MAE-Speech/ | null | null | null | cs.SD cs.LG eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent years have seen remarkable progress in speech emotion recognition
(SER), thanks to advances in deep learning techniques. However, the limited
availability of labeled data remains a significant challenge in the field.
Self-supervised learning has recently emerged as a promising solution to
address this challenge. In this paper, we propose the vector quantized masked
autoencoder for speech (VQ-MAE-S), a self-supervised model that is fine-tuned
to recognize emotions from speech signals. The VQ-MAE-S model is based on a
masked autoencoder (MAE) that operates in the discrete latent space of a
vector-quantized variational autoencoder. Experimental results show that the
proposed VQ-MAE-S model, pre-trained on the VoxCeleb2 dataset and fine-tuned on
emotional speech data, outperforms an MAE working on the raw spectrogram
representation and other state-of-the-art methods in SER.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 16:37:57 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Sadok",
"Samir",
""
],
[
"Leglaive",
"Simon",
""
],
[
"Séguier",
"Renaud",
""
]
] | TITLE: A vector quantized masked autoencoder for speech emotion recognition
ABSTRACT: Recent years have seen remarkable progress in speech emotion recognition
(SER), thanks to advances in deep learning techniques. However, the limited
availability of labeled data remains a significant challenge in the field.
Self-supervised learning has recently emerged as a promising solution to
address this challenge. In this paper, we propose the vector quantized masked
autoencoder for speech (VQ-MAE-S), a self-supervised model that is fine-tuned
to recognize emotions from speech signals. The VQ-MAE-S model is based on a
masked autoencoder (MAE) that operates in the discrete latent space of a
vector-quantized variational autoencoder. Experimental results show that the
proposed VQ-MAE-S model, pre-trained on the VoxCeleb2 dataset and fine-tuned on
emotional speech data, outperforms an MAE working on the raw spectrogram
representation and other state-of-the-art methods in SER.
| no_new_dataset | 0.710616 |
2304.11118 | Angela Castillo | Angela Castillo, Maria Escobar, Guillaume Jeanneret, Albert Pumarola,
Pablo Arbel\'aez, Ali Thabet, Artsiom Sanakoyeu | BoDiffusion: Diffusing Sparse Observations for Full-Body Human Motion
Synthesis | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Mixed reality applications require tracking the user's full-body motion to
enable an immersive experience. However, typical head-mounted devices can only
track head and hand movements, leading to a limited reconstruction of full-body
motion due to variability in lower body configurations. We propose BoDiffusion
-- a generative diffusion model for motion synthesis to tackle this
under-constrained reconstruction problem. We present a time and space
conditioning scheme that allows BoDiffusion to leverage sparse tracking inputs
while generating smooth and realistic full-body motion sequences. To the best
of our knowledge, this is the first approach that uses the reverse diffusion
process to model full-body tracking as a conditional sequence generation task.
We conduct experiments on the large-scale motion-capture dataset AMASS and show
that our approach outperforms the state-of-the-art approaches by a significant
margin in terms of full-body motion realism and joint reconstruction error.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 16:39:05 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Castillo",
"Angela",
""
],
[
"Escobar",
"Maria",
""
],
[
"Jeanneret",
"Guillaume",
""
],
[
"Pumarola",
"Albert",
""
],
[
"Arbeláez",
"Pablo",
""
],
[
"Thabet",
"Ali",
""
],
[
"Sanakoyeu",
"Artsiom",
""
]
] | TITLE: BoDiffusion: Diffusing Sparse Observations for Full-Body Human Motion
Synthesis
ABSTRACT: Mixed reality applications require tracking the user's full-body motion to
enable an immersive experience. However, typical head-mounted devices can only
track head and hand movements, leading to a limited reconstruction of full-body
motion due to variability in lower body configurations. We propose BoDiffusion
-- a generative diffusion model for motion synthesis to tackle this
under-constrained reconstruction problem. We present a time and space
conditioning scheme that allows BoDiffusion to leverage sparse tracking inputs
while generating smooth and realistic full-body motion sequences. To the best
of our knowledge, this is the first approach that uses the reverse diffusion
process to model full-body tracking as a conditional sequence generation task.
We conduct experiments on the large-scale motion-capture dataset AMASS and show
that our approach outperforms the state-of-the-art approaches by a significant
margin in terms of full-body motion realism and joint reconstruction error.
| no_new_dataset | 0.710653 |
2304.11130 | Preslav Nakov | Ashraf Haddad, Najwa Aaraj, Preslav Nakov, Septimiu Fabian Mare | Automated Mapping of CVE Vulnerability Records to MITRE CWE Weaknesses | cybersecurity, MITRE, CVE, CWE | null | null | null | cs.CR cs.AI cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, a proliferation of cyber-security threats and diversity has
been on the rise culminating in an increase in their reporting and analysis. To
counter that, many non-profit organizations have emerged in this domain, such
as MITRE and OSWAP, which have been actively tracking vulnerabilities, and
publishing defense recommendations in standardized formats. As producing data
in such formats manually is very time-consuming, there have been some proposals
to automate the process. Unfortunately, a major obstacle to adopting supervised
machine learning for this problem has been the lack of publicly available
specialized datasets. Here, we aim to bridge this gap. In particular, we focus
on mapping CVE records into MITRE CWE Weaknesses, and we release to the
research community a manually annotated dataset of 4,012 records for this task.
With a human-in-the-loop framework in mind, we approach the problem as a
ranking task and aim to incorporate reinforced learning to make use of the
human feedback in future work. Our experimental results using fine-tuned deep
learning models, namely Sentence-BERT and rankT5, show sizable performance
gains over BM25, BERT, and RoBERTa, which demonstrates the need for an
architecture capable of good semantic understanding for this task.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 19:46:54 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Haddad",
"Ashraf",
""
],
[
"Aaraj",
"Najwa",
""
],
[
"Nakov",
"Preslav",
""
],
[
"Mare",
"Septimiu Fabian",
""
]
] | TITLE: Automated Mapping of CVE Vulnerability Records to MITRE CWE Weaknesses
ABSTRACT: In recent years, a proliferation of cyber-security threats and diversity has
been on the rise culminating in an increase in their reporting and analysis. To
counter that, many non-profit organizations have emerged in this domain, such
as MITRE and OSWAP, which have been actively tracking vulnerabilities, and
publishing defense recommendations in standardized formats. As producing data
in such formats manually is very time-consuming, there have been some proposals
to automate the process. Unfortunately, a major obstacle to adopting supervised
machine learning for this problem has been the lack of publicly available
specialized datasets. Here, we aim to bridge this gap. In particular, we focus
on mapping CVE records into MITRE CWE Weaknesses, and we release to the
research community a manually annotated dataset of 4,012 records for this task.
With a human-in-the-loop framework in mind, we approach the problem as a
ranking task and aim to incorporate reinforced learning to make use of the
human feedback in future work. Our experimental results using fine-tuned deep
learning models, namely Sentence-BERT and rankT5, show sizable performance
gains over BM25, BERT, and RoBERTa, which demonstrates the need for an
architecture capable of good semantic understanding for this task.
| new_dataset | 0.718635 |
2304.11153 | Paul Vicol | Paul Vicol, Zico Kolter, Kevin Swersky | Low-Variance Gradient Estimation in Unrolled Computation Graphs with
ES-Single | null | null | null | null | cs.LG cs.NE stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose an evolution strategies-based algorithm for estimating gradients
in unrolled computation graphs, called ES-Single. Similarly to the
recently-proposed Persistent Evolution Strategies (PES), ES-Single is unbiased,
and overcomes chaos arising from recursive function applications by smoothing
the meta-loss landscape. ES-Single samples a single perturbation per particle,
that is kept fixed over the course of an inner problem (e.g., perturbations are
not re-sampled for each partial unroll). Compared to PES, ES-Single is simpler
to implement and has lower variance: the variance of ES-Single is constant with
respect to the number of truncated unrolls, removing a key barrier in applying
ES to long inner problems using short truncations. We show that ES-Single is
unbiased for quadratic inner problems, and demonstrate empirically that its
variance can be substantially lower than that of PES. ES-Single consistently
outperforms PES on a variety of tasks, including a synthetic benchmark task,
hyperparameter optimization, training recurrent neural networks, and training
learned optimizers.
| [
{
"version": "v1",
"created": "Fri, 21 Apr 2023 17:53:12 GMT"
}
] | 2023-04-24T00:00:00 | [
[
"Vicol",
"Paul",
""
],
[
"Kolter",
"Zico",
""
],
[
"Swersky",
"Kevin",
""
]
] | TITLE: Low-Variance Gradient Estimation in Unrolled Computation Graphs with
ES-Single
ABSTRACT: We propose an evolution strategies-based algorithm for estimating gradients
in unrolled computation graphs, called ES-Single. Similarly to the
recently-proposed Persistent Evolution Strategies (PES), ES-Single is unbiased,
and overcomes chaos arising from recursive function applications by smoothing
the meta-loss landscape. ES-Single samples a single perturbation per particle,
that is kept fixed over the course of an inner problem (e.g., perturbations are
not re-sampled for each partial unroll). Compared to PES, ES-Single is simpler
to implement and has lower variance: the variance of ES-Single is constant with
respect to the number of truncated unrolls, removing a key barrier in applying
ES to long inner problems using short truncations. We show that ES-Single is
unbiased for quadratic inner problems, and demonstrate empirically that its
variance can be substantially lower than that of PES. ES-Single consistently
outperforms PES on a variety of tasks, including a synthetic benchmark task,
hyperparameter optimization, training recurrent neural networks, and training
learned optimizers.
| no_new_dataset | 0.709069 |
2007.03151 | Adel Nabli | Adel Nabli, Margarida Carvalho | Curriculum learning for multilevel budgeted combinatorial problems | NeurIPS 2020, December 2020 | null | null | null | cs.LG cs.GT stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning heuristics for combinatorial optimization problems through graph
neural networks have recently shown promising results on some classic NP-hard
problems. These are single-level optimization problems with only one player.
Multilevel combinatorial optimization problems are their generalization,
encompassing situations with multiple players taking decisions sequentially. By
framing them in a multi-agent reinforcement learning setting, we devise a
value-based method to learn to solve multilevel budgeted combinatorial problems
involving two players in a zero-sum game over a graph. Our framework is based
on a simple curriculum: if an agent knows how to estimate the value of
instances with budgets up to $B$, then solving instances with budget $B+1$ can
be done in polynomial time regardless of the direction of the optimization by
checking the value of every possible afterstate. Thus, in a bottom-up approach,
we generate datasets of heuristically solved instances with increasingly larger
budgets to train our agent. We report results close to optimality on graphs up
to $100$ nodes and a $185 \times$ speedup on average compared to the quickest
exact solver known for the Multilevel Critical Node problem, a max-min-max
trilevel problem that has been shown to be at least $\Sigma_2^p$-hard.
| [
{
"version": "v1",
"created": "Tue, 7 Jul 2020 01:09:37 GMT"
},
{
"version": "v2",
"created": "Mon, 26 Oct 2020 12:23:53 GMT"
}
] | 2023-04-22T00:00:00 | [
[
"Nabli",
"Adel",
""
],
[
"Carvalho",
"Margarida",
""
]
] | TITLE: Curriculum learning for multilevel budgeted combinatorial problems
ABSTRACT: Learning heuristics for combinatorial optimization problems through graph
neural networks have recently shown promising results on some classic NP-hard
problems. These are single-level optimization problems with only one player.
Multilevel combinatorial optimization problems are their generalization,
encompassing situations with multiple players taking decisions sequentially. By
framing them in a multi-agent reinforcement learning setting, we devise a
value-based method to learn to solve multilevel budgeted combinatorial problems
involving two players in a zero-sum game over a graph. Our framework is based
on a simple curriculum: if an agent knows how to estimate the value of
instances with budgets up to $B$, then solving instances with budget $B+1$ can
be done in polynomial time regardless of the direction of the optimization by
checking the value of every possible afterstate. Thus, in a bottom-up approach,
we generate datasets of heuristically solved instances with increasingly larger
budgets to train our agent. We report results close to optimality on graphs up
to $100$ nodes and a $185 \times$ speedup on average compared to the quickest
exact solver known for the Multilevel Critical Node problem, a max-min-max
trilevel problem that has been shown to be at least $\Sigma_2^p$-hard.
| no_new_dataset | 0.707222 |
2004.12231 | Huayu Li | Huayu Li, Xiwen Chen, Haiyu Wu, Zaoyi Chi, Christopher Mann, and
Abolfazl Razi | Deep DIH : Statistically Inferred Reconstruction of Digital In-Line
Holography by Deep Learning | null | null | 10.1109/ACCESS.2020.3036380 | null | eess.IV cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Digital in-line holography is commonly used to reconstruct 3D images from 2D
holograms for microscopic objects. One of the technical challenges that arise
in the signal processing stage is removing the twin image that is caused by the
phase-conjugate wavefront from the recorded holograms. Twin image removal is
typically formulated as a non-linear inverse problem due to the irreversible
scattering process when generating the hologram. Recently, end-to-end deep
learning-based methods have been utilized to reconstruct the object wavefront
(as a surrogate for the 3D structure of the object) directly from a single-shot
in-line digital hologram. However, massive data pairs are required to train
deep learning models for acceptable reconstruction precision. In contrast to
typical image processing problems, well-curated datasets for in-line digital
holography does not exist. Also, the trained model highly influenced by the
morphological properties of the object and hence can vary for different
applications. Therefore, data collection can be prohibitively cumbersome in
practice as a major hindrance to using deep learning for digital holography. In
this paper, we proposed a novel implementation of autoencoder-based deep
learning architecture for single-shot hologram reconstruction solely based on
the current sample without the need for massive datasets to train the model.
The simulations results demonstrate the superior performance of the proposed
method compared to the state of the art single-shot compressive digital in-line
hologram reconstruction method.
| [
{
"version": "v1",
"created": "Sat, 25 Apr 2020 20:39:25 GMT"
},
{
"version": "v2",
"created": "Wed, 24 Jun 2020 22:08:02 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Li",
"Huayu",
""
],
[
"Chen",
"Xiwen",
""
],
[
"Wu",
"Haiyu",
""
],
[
"Chi",
"Zaoyi",
""
],
[
"Mann",
"Christopher",
""
],
[
"Razi",
"Abolfazl",
""
]
] | TITLE: Deep DIH : Statistically Inferred Reconstruction of Digital In-Line
Holography by Deep Learning
ABSTRACT: Digital in-line holography is commonly used to reconstruct 3D images from 2D
holograms for microscopic objects. One of the technical challenges that arise
in the signal processing stage is removing the twin image that is caused by the
phase-conjugate wavefront from the recorded holograms. Twin image removal is
typically formulated as a non-linear inverse problem due to the irreversible
scattering process when generating the hologram. Recently, end-to-end deep
learning-based methods have been utilized to reconstruct the object wavefront
(as a surrogate for the 3D structure of the object) directly from a single-shot
in-line digital hologram. However, massive data pairs are required to train
deep learning models for acceptable reconstruction precision. In contrast to
typical image processing problems, well-curated datasets for in-line digital
holography does not exist. Also, the trained model highly influenced by the
morphological properties of the object and hence can vary for different
applications. Therefore, data collection can be prohibitively cumbersome in
practice as a major hindrance to using deep learning for digital holography. In
this paper, we proposed a novel implementation of autoencoder-based deep
learning architecture for single-shot hologram reconstruction solely based on
the current sample without the need for massive datasets to train the model.
The simulations results demonstrate the superior performance of the proposed
method compared to the state of the art single-shot compressive digital in-line
hologram reconstruction method.
| no_new_dataset | 0.711212 |
2007.01975 | Ricardo Bigolin Lanfredi | Ricardo Bigolin Lanfredi, Joyce D. Schroeder, Clement Vachet, Tolga
Tasdizen | Interpretation of Disease Evidence for Medical Images Using Adversarial
Deformation Fields | Presented at MICCAI 2020 | null | 10.1007/978-3-030-59713-9_71 | null | eess.IV cs.CV cs.GR q-bio.QM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The high complexity of deep learning models is associated with the difficulty
of explaining what evidence they recognize as correlating with specific disease
labels. This information is critical for building trust in models and finding
their biases. Until now, automated deep learning visualization solutions have
identified regions of images used by classifiers, but these solutions are too
coarse, too noisy, or have a limited representation of the way images can
change. We propose a novel method for formulating and presenting spatial
explanations of disease evidence, called deformation field interpretation with
generative adversarial networks (DeFI-GAN). An adversarially trained generator
produces deformation fields that modify images of diseased patients to resemble
images of healthy patients. We validate the method studying chronic obstructive
pulmonary disease (COPD) evidence in chest x-rays (CXRs) and Alzheimer's
disease (AD) evidence in brain MRIs. When extracting disease evidence in
longitudinal data, we show compelling results against a baseline producing
difference maps. DeFI-GAN also highlights disease biomarkers not found by
previous methods and potential biases that may help in investigations of the
dataset and of the adopted learning methods.
| [
{
"version": "v1",
"created": "Sat, 4 Jul 2020 00:51:54 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 02:05:48 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Lanfredi",
"Ricardo Bigolin",
""
],
[
"Schroeder",
"Joyce D.",
""
],
[
"Vachet",
"Clement",
""
],
[
"Tasdizen",
"Tolga",
""
]
] | TITLE: Interpretation of Disease Evidence for Medical Images Using Adversarial
Deformation Fields
ABSTRACT: The high complexity of deep learning models is associated with the difficulty
of explaining what evidence they recognize as correlating with specific disease
labels. This information is critical for building trust in models and finding
their biases. Until now, automated deep learning visualization solutions have
identified regions of images used by classifiers, but these solutions are too
coarse, too noisy, or have a limited representation of the way images can
change. We propose a novel method for formulating and presenting spatial
explanations of disease evidence, called deformation field interpretation with
generative adversarial networks (DeFI-GAN). An adversarially trained generator
produces deformation fields that modify images of diseased patients to resemble
images of healthy patients. We validate the method studying chronic obstructive
pulmonary disease (COPD) evidence in chest x-rays (CXRs) and Alzheimer's
disease (AD) evidence in brain MRIs. When extracting disease evidence in
longitudinal data, we show compelling results against a baseline producing
difference maps. DeFI-GAN also highlights disease biomarkers not found by
previous methods and potential biases that may help in investigations of the
dataset and of the adopted learning methods.
| no_new_dataset | 0.709787 |
2103.06638 | Nicola Strisciuglio | Mar\'ia Leyva-Vallina, Nicola Strisciuglio, Nicolai Petkov | Generalized Contrastive Optimization of Siamese Networks for Place
Recognition | Published at CVPR2023 as arXiv:2303.11739 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual place recognition is a challenging task in computer vision and a key
component of camera-based localization and navigation systems. Recently,
Convolutional Neural Networks (CNNs) achieved high results and good
generalization capabilities. They are usually trained using pairs or triplets
of images labeled as either similar or dissimilar, in a binary fashion. In
practice, the similarity between two images is not binary, but continuous.
Furthermore, training these CNNs is computationally complex and involves costly
pair and triplet mining strategies. We propose a Generalized Contrastive loss
(GCL) function that relies on image similarity as a continuous measure, and use
it to train a siamese CNN. Furthermore, we present three techniques for
automatic annotation of image pairs with labels indicating their degree of
similarity, and deploy them to re-annotate the MSLS, TB-Places, and 7Scenes
datasets. We demonstrate that siamese CNNs trained using the GCL function and
the improved annotations consistently outperform their binary counterparts. Our
models trained on MSLS outperform the state-of-the-art methods, including
NetVLAD, NetVLAD-SARE, AP-GeM and Patch-NetVLAD, and generalize well on the
Pittsburgh30k, Tokyo 24/7, RobotCar Seasons v2 and Extended CMU Seasons
datasets. Furthermore, training a siamese network using the GCL function does
not require complex pair mining. We release the source code at
https://github.com/marialeyvallina/generalized_contrastive_loss.
| [
{
"version": "v1",
"created": "Thu, 11 Mar 2021 12:32:05 GMT"
},
{
"version": "v2",
"created": "Mon, 24 Jan 2022 10:02:49 GMT"
},
{
"version": "v3",
"created": "Mon, 9 May 2022 12:37:22 GMT"
},
{
"version": "v4",
"created": "Thu, 20 Apr 2023 09:24:25 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Leyva-Vallina",
"María",
""
],
[
"Strisciuglio",
"Nicola",
""
],
[
"Petkov",
"Nicolai",
""
]
] | TITLE: Generalized Contrastive Optimization of Siamese Networks for Place
Recognition
ABSTRACT: Visual place recognition is a challenging task in computer vision and a key
component of camera-based localization and navigation systems. Recently,
Convolutional Neural Networks (CNNs) achieved high results and good
generalization capabilities. They are usually trained using pairs or triplets
of images labeled as either similar or dissimilar, in a binary fashion. In
practice, the similarity between two images is not binary, but continuous.
Furthermore, training these CNNs is computationally complex and involves costly
pair and triplet mining strategies. We propose a Generalized Contrastive loss
(GCL) function that relies on image similarity as a continuous measure, and use
it to train a siamese CNN. Furthermore, we present three techniques for
automatic annotation of image pairs with labels indicating their degree of
similarity, and deploy them to re-annotate the MSLS, TB-Places, and 7Scenes
datasets. We demonstrate that siamese CNNs trained using the GCL function and
the improved annotations consistently outperform their binary counterparts. Our
models trained on MSLS outperform the state-of-the-art methods, including
NetVLAD, NetVLAD-SARE, AP-GeM and Patch-NetVLAD, and generalize well on the
Pittsburgh30k, Tokyo 24/7, RobotCar Seasons v2 and Extended CMU Seasons
datasets. Furthermore, training a siamese network using the GCL function does
not require complex pair mining. We release the source code at
https://github.com/marialeyvallina/generalized_contrastive_loss.
| no_new_dataset | 0.71222 |
2112.11716 | Ricardo Bigolin Lanfredi | Ricardo Bigolin Lanfredi, Ambuj Arora, Trafton Drew, Joyce D.
Schroeder, Tolga Tasdizen | Comparing radiologists' gaze and saliency maps generated by
interpretability methods for chest x-rays | This paper was presented as an Extended Abstract at the Gaze Meets ML
2022 Workshop, a NeurIPS 2022 workshop | null | null | null | cs.CV eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The interpretability of medical image analysis models is considered a key
research field. We use a dataset of eye-tracking data from five radiologists to
compare the outputs of interpretability methods and the heatmaps representing
where radiologists looked. We conduct a class-independent analysis of the
saliency maps generated by two methods selected from the literature: Grad-CAM
and attention maps from an attention-gated model. For the comparison, we use
shuffled metrics, which avoid biases from fixation locations. We achieve scores
comparable to an interobserver baseline in one shuffled metric, highlighting
the potential of saliency maps from Grad-CAM to mimic a radiologist's attention
over an image. We also divide the dataset into subsets to evaluate in which
cases similarities are higher.
| [
{
"version": "v1",
"created": "Wed, 22 Dec 2021 07:57:18 GMT"
},
{
"version": "v2",
"created": "Fri, 24 Jun 2022 23:22:49 GMT"
},
{
"version": "v3",
"created": "Thu, 20 Apr 2023 01:58:51 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Lanfredi",
"Ricardo Bigolin",
""
],
[
"Arora",
"Ambuj",
""
],
[
"Drew",
"Trafton",
""
],
[
"Schroeder",
"Joyce D.",
""
],
[
"Tasdizen",
"Tolga",
""
]
] | TITLE: Comparing radiologists' gaze and saliency maps generated by
interpretability methods for chest x-rays
ABSTRACT: The interpretability of medical image analysis models is considered a key
research field. We use a dataset of eye-tracking data from five radiologists to
compare the outputs of interpretability methods and the heatmaps representing
where radiologists looked. We conduct a class-independent analysis of the
saliency maps generated by two methods selected from the literature: Grad-CAM
and attention maps from an attention-gated model. For the comparison, we use
shuffled metrics, which avoid biases from fixation locations. We achieve scores
comparable to an interobserver baseline in one shuffled metric, highlighting
the potential of saliency maps from Grad-CAM to mimic a radiologist's attention
over an image. We also divide the dataset into subsets to evaluate in which
cases similarities are higher.
| no_new_dataset | 0.701291 |
2112.12591 | Manel Abdellatif | Zohreh Aghababaeyan, Manel Abdellatif, Lionel Briand, Ramesh S, and
Mojtaba Bagherzadeh | Black-Box Testing of Deep Neural Networks Through Test Case Diversity | null | IEEE Transactions on Software Engineering (TSE) (2023) 1-26 | 10.1109/TSE.2023.3243522 | null | cs.SE cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep Neural Networks (DNNs) have been extensively used in many areas
including image processing, medical diagnostics, and autonomous driving.
However, DNNs can exhibit erroneous behaviours that may lead to critical
errors, especially when used in safety-critical systems. Inspired by testing
techniques for traditional software systems, researchers have proposed neuron
coverage criteria, as an analogy to source code coverage, to guide the testing
of DNN models. Despite very active research on DNN coverage, several recent
studies have questioned the usefulness of such criteria in guiding DNN testing.
Further, from a practical standpoint, these criteria are white-box as they
require access to the internals or training data of DNN models, which is in
many contexts not feasible or convenient. In this paper, we investigate
black-box input diversity metrics as an alternative to white-box coverage
criteria. To this end, we first select and adapt three diversity metrics and
study, in a controlled manner, their capacity to measure actual diversity in
input sets. We then analyse their statistical association with fault detection
using four datasets and five DNN models. We further compare diversity with
state-of-the-art white-box coverage criteria. Our experiments show that relying
on the diversity of image features embedded in test input sets is a more
reliable indicator than coverage criteria to effectively guide the testing of
DNNs. Indeed, we found that one of our selected black-box diversity metrics far
outperforms existing coverage criteria in terms of fault-revealing capability
and computational time. Results also confirm the suspicions that
state-of-the-art coverage metrics are not adequate to guide the construction of
test input sets to detect as many faults as possible with natural inputs.
| [
{
"version": "v1",
"created": "Mon, 20 Dec 2021 20:12:53 GMT"
},
{
"version": "v2",
"created": "Tue, 18 Jan 2022 18:32:21 GMT"
},
{
"version": "v3",
"created": "Tue, 21 Jun 2022 11:24:45 GMT"
},
{
"version": "v4",
"created": "Mon, 5 Dec 2022 16:15:52 GMT"
},
{
"version": "v5",
"created": "Mon, 6 Feb 2023 19:02:57 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Aghababaeyan",
"Zohreh",
""
],
[
"Abdellatif",
"Manel",
""
],
[
"Briand",
"Lionel",
""
],
[
"S",
"Ramesh",
""
],
[
"Bagherzadeh",
"Mojtaba",
""
]
] | TITLE: Black-Box Testing of Deep Neural Networks Through Test Case Diversity
ABSTRACT: Deep Neural Networks (DNNs) have been extensively used in many areas
including image processing, medical diagnostics, and autonomous driving.
However, DNNs can exhibit erroneous behaviours that may lead to critical
errors, especially when used in safety-critical systems. Inspired by testing
techniques for traditional software systems, researchers have proposed neuron
coverage criteria, as an analogy to source code coverage, to guide the testing
of DNN models. Despite very active research on DNN coverage, several recent
studies have questioned the usefulness of such criteria in guiding DNN testing.
Further, from a practical standpoint, these criteria are white-box as they
require access to the internals or training data of DNN models, which is in
many contexts not feasible or convenient. In this paper, we investigate
black-box input diversity metrics as an alternative to white-box coverage
criteria. To this end, we first select and adapt three diversity metrics and
study, in a controlled manner, their capacity to measure actual diversity in
input sets. We then analyse their statistical association with fault detection
using four datasets and five DNN models. We further compare diversity with
state-of-the-art white-box coverage criteria. Our experiments show that relying
on the diversity of image features embedded in test input sets is a more
reliable indicator than coverage criteria to effectively guide the testing of
DNNs. Indeed, we found that one of our selected black-box diversity metrics far
outperforms existing coverage criteria in terms of fault-revealing capability
and computational time. Results also confirm the suspicions that
state-of-the-art coverage metrics are not adequate to guide the construction of
test input sets to detect as many faults as possible with natural inputs.
| no_new_dataset | 0.708452 |
2203.03397 | Xieyuanli Chen | Junyi Ma and Jun Zhang and Jintao Xu and Rui Ai and Weihao Gu and
Xieyuanli Chen | OverlapTransformer: An Efficient and Rotation-Invariant Transformer
Network for LiDAR-Based Place Recognition | Accepted by RAL/IROS 2022 | null | 10.1109/LRA.2022.3178797 | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Place recognition is an important capability for autonomously navigating
vehicles operating in complex environments and under changing conditions. It is
a key component for tasks such as loop closing in SLAM or global localization.
In this paper, we address the problem of place recognition based on 3D LiDAR
scans recorded by an autonomous vehicle. We propose a novel lightweight neural
network exploiting the range image representation of LiDAR sensors to achieve
fast execution with less than 2 ms per frame. We design a yaw-angle-invariant
architecture exploiting a transformer network, which boosts the place
recognition performance of our method. We evaluate our approach on the KITTI
and Ford Campus datasets. The experimental results show that our method can
effectively detect loop closures compared to the state-of-the-art methods and
generalizes well across different environments. To evaluate long-term place
recognition performance, we provide a novel dataset containing LiDAR sequences
recorded by a mobile robot in repetitive places at different times. The
implementation of our method and dataset are released here:
https://github.com/haomo-ai/OverlapTransformer
| [
{
"version": "v1",
"created": "Mon, 7 Mar 2022 13:49:09 GMT"
},
{
"version": "v2",
"created": "Fri, 20 May 2022 17:37:05 GMT"
},
{
"version": "v3",
"created": "Fri, 27 May 2022 13:41:27 GMT"
},
{
"version": "v4",
"created": "Thu, 20 Apr 2023 02:40:33 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Ma",
"Junyi",
""
],
[
"Zhang",
"Jun",
""
],
[
"Xu",
"Jintao",
""
],
[
"Ai",
"Rui",
""
],
[
"Gu",
"Weihao",
""
],
[
"Chen",
"Xieyuanli",
""
]
] | TITLE: OverlapTransformer: An Efficient and Rotation-Invariant Transformer
Network for LiDAR-Based Place Recognition
ABSTRACT: Place recognition is an important capability for autonomously navigating
vehicles operating in complex environments and under changing conditions. It is
a key component for tasks such as loop closing in SLAM or global localization.
In this paper, we address the problem of place recognition based on 3D LiDAR
scans recorded by an autonomous vehicle. We propose a novel lightweight neural
network exploiting the range image representation of LiDAR sensors to achieve
fast execution with less than 2 ms per frame. We design a yaw-angle-invariant
architecture exploiting a transformer network, which boosts the place
recognition performance of our method. We evaluate our approach on the KITTI
and Ford Campus datasets. The experimental results show that our method can
effectively detect loop closures compared to the state-of-the-art methods and
generalizes well across different environments. To evaluate long-term place
recognition performance, we provide a novel dataset containing LiDAR sequences
recorded by a mobile robot in repetitive places at different times. The
implementation of our method and dataset are released here:
https://github.com/haomo-ai/OverlapTransformer
| new_dataset | 0.714634 |
2203.09135 | Qiang Zhai | Jianwei Zhao and Qiang Zhai and Pengbo Zhao and Rui Huang and Hong
Cheng | Co-visual pattern augmented generative transformer learning for
automobile geo-localization | 21pages | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Geolocation is a fundamental component of route planning and navigation for
unmanned vehicles, but GNSS-based geolocation fails under denial-of-service
conditions. Cross-view geo-localization (CVGL), which aims to estimate the
geographical location of the ground-level camera by matching against enormous
geo-tagged aerial (\emph{e.g.}, satellite) images, has received lots of
attention but remains extremely challenging due to the drastic appearance
differences across aerial-ground views. In existing methods, global
representations of different views are extracted primarily using Siamese-like
architectures, but their interactive benefits are seldom taken into account. In
this paper, we present a novel approach using cross-view knowledge generative
techniques in combination with transformers, namely mutual generative
transformer learning (MGTL), for CVGL. Specifically, by taking the initial
representations produced by the backbone network, MGTL develops two separate
generative sub-modules -- one for aerial-aware knowledge generation from
ground-view semantics and vice versa -- and fully exploits the entirely mutual
benefits through the attention mechanism. Moreover, to better capture the
co-visual relationships between aerial and ground views, we introduce a
cascaded attention masking algorithm to further boost accuracy. Extensive
experiments on challenging public benchmarks, \emph{i.e.}, {CVACT} and {CVUSA},
demonstrate the effectiveness of the proposed method which sets new records
compared with the existing state-of-the-art models.
| [
{
"version": "v1",
"created": "Thu, 17 Mar 2022 07:29:02 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 12:49:27 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Zhao",
"Jianwei",
""
],
[
"Zhai",
"Qiang",
""
],
[
"Zhao",
"Pengbo",
""
],
[
"Huang",
"Rui",
""
],
[
"Cheng",
"Hong",
""
]
] | TITLE: Co-visual pattern augmented generative transformer learning for
automobile geo-localization
ABSTRACT: Geolocation is a fundamental component of route planning and navigation for
unmanned vehicles, but GNSS-based geolocation fails under denial-of-service
conditions. Cross-view geo-localization (CVGL), which aims to estimate the
geographical location of the ground-level camera by matching against enormous
geo-tagged aerial (\emph{e.g.}, satellite) images, has received lots of
attention but remains extremely challenging due to the drastic appearance
differences across aerial-ground views. In existing methods, global
representations of different views are extracted primarily using Siamese-like
architectures, but their interactive benefits are seldom taken into account. In
this paper, we present a novel approach using cross-view knowledge generative
techniques in combination with transformers, namely mutual generative
transformer learning (MGTL), for CVGL. Specifically, by taking the initial
representations produced by the backbone network, MGTL develops two separate
generative sub-modules -- one for aerial-aware knowledge generation from
ground-view semantics and vice versa -- and fully exploits the entirely mutual
benefits through the attention mechanism. Moreover, to better capture the
co-visual relationships between aerial and ground views, we introduce a
cascaded attention masking algorithm to further boost accuracy. Extensive
experiments on challenging public benchmarks, \emph{i.e.}, {CVACT} and {CVUSA},
demonstrate the effectiveness of the proposed method which sets new records
compared with the existing state-of-the-art models.
| no_new_dataset | 0.709799 |
2204.01218 | Tiantian Wang | Tiantian Wang, Nikolaos Sarafianos, Ming-Hsuan Yang, Tony Tung | Neural Rendering of Humans in Novel View and Pose from Monocular Video | 10 pages | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We introduce a new method that generates photo-realistic humans under novel
views and poses given a monocular video as input. Despite the significant
progress recently on this topic, with several methods exploring shared
canonical neural radiance fields in dynamic scene scenarios, learning a
user-controlled model for unseen poses remains a challenging task. To tackle
this problem, we introduce an effective method to a) integrate observations
across several frames and b) encode the appearance at each individual frame. We
accomplish this by utilizing both the human pose that models the body shape as
well as point clouds that partially cover the human as input. Our approach
simultaneously learns a shared set of latent codes anchored to the human pose
among several frames, and an appearance-dependent code anchored to incomplete
point clouds generated by each frame and its predicted depth. The former human
pose-based code models the shape of the performer whereas the latter point
cloud-based code predicts fine-level details and reasons about missing
structures at the unseen poses. To further recover non-visible regions in query
frames, we employ a temporal transformer to integrate features of points in
query frames and tracked body points from automatically-selected key frames.
Experiments on various sequences of dynamic humans from different datasets
including ZJU-MoCap show that our method significantly outperforms existing
approaches under unseen poses and novel views given monocular videos as input.
| [
{
"version": "v1",
"created": "Mon, 4 Apr 2022 03:09:20 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 04:08:04 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Wang",
"Tiantian",
""
],
[
"Sarafianos",
"Nikolaos",
""
],
[
"Yang",
"Ming-Hsuan",
""
],
[
"Tung",
"Tony",
""
]
] | TITLE: Neural Rendering of Humans in Novel View and Pose from Monocular Video
ABSTRACT: We introduce a new method that generates photo-realistic humans under novel
views and poses given a monocular video as input. Despite the significant
progress recently on this topic, with several methods exploring shared
canonical neural radiance fields in dynamic scene scenarios, learning a
user-controlled model for unseen poses remains a challenging task. To tackle
this problem, we introduce an effective method to a) integrate observations
across several frames and b) encode the appearance at each individual frame. We
accomplish this by utilizing both the human pose that models the body shape as
well as point clouds that partially cover the human as input. Our approach
simultaneously learns a shared set of latent codes anchored to the human pose
among several frames, and an appearance-dependent code anchored to incomplete
point clouds generated by each frame and its predicted depth. The former human
pose-based code models the shape of the performer whereas the latter point
cloud-based code predicts fine-level details and reasons about missing
structures at the unseen poses. To further recover non-visible regions in query
frames, we employ a temporal transformer to integrate features of points in
query frames and tracked body points from automatically-selected key frames.
Experiments on various sequences of dynamic humans from different datasets
including ZJU-MoCap show that our method significantly outperforms existing
approaches under unseen poses and novel views given monocular videos as input.
| no_new_dataset | 0.710666 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.