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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2204.03245 | Jiashun Suo | Jiashun Suo, Tianyi Wang, Xingzhou Zhang, Haiyang Chen, Wei Zhou,
Weisong Shi | HIT-UAV: A high-altitude infrared thermal dataset for Unmanned Aerial
Vehicle-based object detection | null | Sci Data 10, 227 (2023) | 10.1038/s41597-023-02066-6 | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present the HIT-UAV dataset, a high-altitude infrared thermal dataset for
object detection applications on Unmanned Aerial Vehicles (UAVs). The dataset
comprises 2,898 infrared thermal images extracted from 43,470 frames in
hundreds of videos captured by UAVs in various scenarios including schools,
parking lots, roads, and playgrounds. Moreover, the HIT-UAV provides essential
flight data for each image, such as flight altitude, camera perspective, date,
and daylight intensity. For each image, we have manually annotated object
instances with bounding boxes of two types (oriented and standard) to tackle
the challenge of significant overlap of object instances in aerial images. To
the best of our knowledge, the HIT-UAV is the first publicly available
high-altitude UAV-based infrared thermal dataset for detecting persons and
vehicles. We have trained and evaluated well-established object detection
algorithms on the HIT-UAV. Our results demonstrate that the detection
algorithms perform exceptionally well on the HIT-UAV compared to visual light
datasets since infrared thermal images do not contain significant irrelevant
information about objects. We believe that the HIT-UAV will contribute to
various UAV-based applications and researches. The dataset is freely available
at https://github.com/suojiashun/HIT-UAV-Infrared-Thermal-Dataset.
| [
{
"version": "v1",
"created": "Thu, 7 Apr 2022 06:23:02 GMT"
},
{
"version": "v2",
"created": "Fri, 31 Mar 2023 11:04:55 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Suo",
"Jiashun",
""
],
[
"Wang",
"Tianyi",
""
],
[
"Zhang",
"Xingzhou",
""
],
[
"Chen",
"Haiyang",
""
],
[
"Zhou",
"Wei",
""
],
[
"Shi",
"Weisong",
""
]
] | TITLE: HIT-UAV: A high-altitude infrared thermal dataset for Unmanned Aerial
Vehicle-based object detection
ABSTRACT: We present the HIT-UAV dataset, a high-altitude infrared thermal dataset for
object detection applications on Unmanned Aerial Vehicles (UAVs). The dataset
comprises 2,898 infrared thermal images extracted from 43,470 frames in
hundreds of videos captured by UAVs in various scenarios including schools,
parking lots, roads, and playgrounds. Moreover, the HIT-UAV provides essential
flight data for each image, such as flight altitude, camera perspective, date,
and daylight intensity. For each image, we have manually annotated object
instances with bounding boxes of two types (oriented and standard) to tackle
the challenge of significant overlap of object instances in aerial images. To
the best of our knowledge, the HIT-UAV is the first publicly available
high-altitude UAV-based infrared thermal dataset for detecting persons and
vehicles. We have trained and evaluated well-established object detection
algorithms on the HIT-UAV. Our results demonstrate that the detection
algorithms perform exceptionally well on the HIT-UAV compared to visual light
datasets since infrared thermal images do not contain significant irrelevant
information about objects. We believe that the HIT-UAV will contribute to
various UAV-based applications and researches. The dataset is freely available
at https://github.com/suojiashun/HIT-UAV-Infrared-Thermal-Dataset.
|
2204.08461 | James Brock | James Brock, Zahraa S. Abdallah | Investigating Temporal Convolutional Neural Networks for Satellite Image
Time Series Classification: A survey | 24 pages (not inc. 6 pages for appendices and references), 8 figures,
15 tables. Submitted for publishing | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Satellite Image Time Series (SITS) of the Earth's surface provide detailed
land cover maps, with their quality in the spatial and temporal dimensions
consistently improving. These image time series are integral for developing
systems that aim to produce accurate, up-to-date land cover maps of the Earth's
surface. Applications are wide-ranging, with notable examples including
ecosystem mapping, vegetation process monitoring and anthropogenic land-use
change tracking. Recently proposed methods for SITS classification have
demonstrated respectable merit, but these methods tend to lack native
mechanisms that exploit the temporal dimension of the data; commonly resulting
in extensive data pre-processing contributing to prohibitively long training
times. To overcome these shortcomings, Temporal CNNs have recently been
employed for SITS classification tasks with encouraging results. This paper
seeks to survey this method against a plethora of other contemporary methods
for SITS classification to validate the existing findings in recent literature.
Comprehensive experiments are carried out on two benchmark SITS datasets with
the results demonstrating that Temporal CNNs display a superior performance to
the comparative benchmark algorithms across both studied datasets, achieving
accuracies of 95.0\% and 87.3\% respectively. Investigations into the Temporal
CNN architecture also highlighted the non-trivial task of optimising the model
for a new dataset.
| [
{
"version": "v1",
"created": "Wed, 13 Apr 2022 14:08:14 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 13:58:18 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Brock",
"James",
""
],
[
"Abdallah",
"Zahraa S.",
""
]
] | TITLE: Investigating Temporal Convolutional Neural Networks for Satellite Image
Time Series Classification: A survey
ABSTRACT: Satellite Image Time Series (SITS) of the Earth's surface provide detailed
land cover maps, with their quality in the spatial and temporal dimensions
consistently improving. These image time series are integral for developing
systems that aim to produce accurate, up-to-date land cover maps of the Earth's
surface. Applications are wide-ranging, with notable examples including
ecosystem mapping, vegetation process monitoring and anthropogenic land-use
change tracking. Recently proposed methods for SITS classification have
demonstrated respectable merit, but these methods tend to lack native
mechanisms that exploit the temporal dimension of the data; commonly resulting
in extensive data pre-processing contributing to prohibitively long training
times. To overcome these shortcomings, Temporal CNNs have recently been
employed for SITS classification tasks with encouraging results. This paper
seeks to survey this method against a plethora of other contemporary methods
for SITS classification to validate the existing findings in recent literature.
Comprehensive experiments are carried out on two benchmark SITS datasets with
the results demonstrating that Temporal CNNs display a superior performance to
the comparative benchmark algorithms across both studied datasets, achieving
accuracies of 95.0\% and 87.3\% respectively. Investigations into the Temporal
CNN architecture also highlighted the non-trivial task of optimising the model
for a new dataset.
|
2205.02719 | Yujia Wang | Yujia Wang, Lu Lin, Jinghui Chen | Communication-Efficient Adaptive Federated Learning | Updated version. A previous version is accepted by ICML 2022 (37
pages, 7 figures, 3 tables) | null | null | null | cs.LG cs.AI cs.DC math.OC stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated learning is a machine learning training paradigm that enables
clients to jointly train models without sharing their own localized data.
However, the implementation of federated learning in practice still faces
numerous challenges, such as the large communication overhead due to the
repetitive server-client synchronization and the lack of adaptivity by
SGD-based model updates. Despite that various methods have been proposed for
reducing the communication cost by gradient compression or quantization, and
the federated versions of adaptive optimizers such as FedAdam are proposed to
add more adaptivity, the current federated learning framework still cannot
solve the aforementioned challenges all at once. In this paper, we propose a
novel communication-efficient adaptive federated learning method (FedCAMS) with
theoretical convergence guarantees. We show that in the nonconvex stochastic
optimization setting, our proposed FedCAMS achieves the same convergence rate
of $O(\frac{1}{\sqrt{TKm}})$ as its non-compressed counterparts. Extensive
experiments on various benchmarks verify our theoretical analysis.
| [
{
"version": "v1",
"created": "Thu, 5 May 2022 15:47:04 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Jun 2022 00:28:37 GMT"
},
{
"version": "v3",
"created": "Wed, 19 Apr 2023 18:20:06 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Wang",
"Yujia",
""
],
[
"Lin",
"Lu",
""
],
[
"Chen",
"Jinghui",
""
]
] | TITLE: Communication-Efficient Adaptive Federated Learning
ABSTRACT: Federated learning is a machine learning training paradigm that enables
clients to jointly train models without sharing their own localized data.
However, the implementation of federated learning in practice still faces
numerous challenges, such as the large communication overhead due to the
repetitive server-client synchronization and the lack of adaptivity by
SGD-based model updates. Despite that various methods have been proposed for
reducing the communication cost by gradient compression or quantization, and
the federated versions of adaptive optimizers such as FedAdam are proposed to
add more adaptivity, the current federated learning framework still cannot
solve the aforementioned challenges all at once. In this paper, we propose a
novel communication-efficient adaptive federated learning method (FedCAMS) with
theoretical convergence guarantees. We show that in the nonconvex stochastic
optimization setting, our proposed FedCAMS achieves the same convergence rate
of $O(\frac{1}{\sqrt{TKm}})$ as its non-compressed counterparts. Extensive
experiments on various benchmarks verify our theoretical analysis.
|
2205.11039 | Xueyuan Lin | Xueyuan Lin, Haihong E, Gengxian Zhou, Tianyi Hu, Li Ningyuan, Mingzhi
Sun, Haoran Luo | FLEX: Feature-Logic Embedding Framework for CompleX Knowledge Graph
Reasoning | null | null | null | null | cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Current best performing models for knowledge graph reasoning (KGR) introduce
geometry objects or probabilistic distributions to embed entities and
first-order logical (FOL) queries into low-dimensional vector spaces. They can
be summarized as a center-size framework (point/box/cone, Beta/Gaussian
distribution, etc.). However, they have limited logical reasoning ability. And
it is difficult to generalize to various features, because the center and size
are one-to-one constrained, unable to have multiple centers or sizes. To
address these challenges, we instead propose a novel KGR framework named
Feature-Logic Embedding framework, FLEX, which is the first KGR framework that
can not only TRULY handle all FOL operations including conjunction,
disjunction, negation and so on, but also support various feature spaces.
Specifically, the logic part of feature-logic framework is based on vector
logic, which naturally models all FOL operations. Experiments demonstrate that
FLEX significantly outperforms existing state-of-the-art methods on benchmark
datasets.
| [
{
"version": "v1",
"created": "Mon, 23 May 2022 05:05:27 GMT"
},
{
"version": "v2",
"created": "Fri, 23 Sep 2022 12:53:14 GMT"
},
{
"version": "v3",
"created": "Thu, 20 Apr 2023 10:54:02 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Lin",
"Xueyuan",
""
],
[
"E",
"Haihong",
""
],
[
"Zhou",
"Gengxian",
""
],
[
"Hu",
"Tianyi",
""
],
[
"Ningyuan",
"Li",
""
],
[
"Sun",
"Mingzhi",
""
],
[
"Luo",
"Haoran",
""
]
] | TITLE: FLEX: Feature-Logic Embedding Framework for CompleX Knowledge Graph
Reasoning
ABSTRACT: Current best performing models for knowledge graph reasoning (KGR) introduce
geometry objects or probabilistic distributions to embed entities and
first-order logical (FOL) queries into low-dimensional vector spaces. They can
be summarized as a center-size framework (point/box/cone, Beta/Gaussian
distribution, etc.). However, they have limited logical reasoning ability. And
it is difficult to generalize to various features, because the center and size
are one-to-one constrained, unable to have multiple centers or sizes. To
address these challenges, we instead propose a novel KGR framework named
Feature-Logic Embedding framework, FLEX, which is the first KGR framework that
can not only TRULY handle all FOL operations including conjunction,
disjunction, negation and so on, but also support various feature spaces.
Specifically, the logic part of feature-logic framework is based on vector
logic, which naturally models all FOL operations. Experiments demonstrate that
FLEX significantly outperforms existing state-of-the-art methods on benchmark
datasets.
|
2206.00686 | Huancheng Chen | Huancheng Chen and Haris Vikalo | Federated Learning in Non-IID Settings Aided by Differentially Private
Synthetic Data | To appear in CVPR2023 Workshop Federated Learning in Computer Vision | null | null | null | cs.LG cs.CR cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated learning (FL) is a privacy-promoting framework that enables
potentially large number of clients to collaboratively train machine learning
models. In a FL system, a server coordinates the collaboration by collecting
and aggregating clients' model updates while the clients' data remains local
and private. A major challenge in federated learning arises when the local data
is heterogeneous -- the setting in which performance of the learned global
model may deteriorate significantly compared to the scenario where the data is
identically distributed across the clients. In this paper we propose FedDPMS
(Federated Differentially Private Means Sharing), an FL algorithm in which
clients deploy variational auto-encoders to augment local datasets with data
synthesized using differentially private means of latent data representations
communicated by a trusted server. Such augmentation ameliorates effects of data
heterogeneity across the clients without compromising privacy. Our experiments
on deep image classification tasks demonstrate that FedDPMS outperforms
competing state-of-the-art FL methods specifically designed for heterogeneous
data settings.
| [
{
"version": "v1",
"created": "Wed, 1 Jun 2022 18:00:48 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 01:50:34 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Chen",
"Huancheng",
""
],
[
"Vikalo",
"Haris",
""
]
] | TITLE: Federated Learning in Non-IID Settings Aided by Differentially Private
Synthetic Data
ABSTRACT: Federated learning (FL) is a privacy-promoting framework that enables
potentially large number of clients to collaboratively train machine learning
models. In a FL system, a server coordinates the collaboration by collecting
and aggregating clients' model updates while the clients' data remains local
and private. A major challenge in federated learning arises when the local data
is heterogeneous -- the setting in which performance of the learned global
model may deteriorate significantly compared to the scenario where the data is
identically distributed across the clients. In this paper we propose FedDPMS
(Federated Differentially Private Means Sharing), an FL algorithm in which
clients deploy variational auto-encoders to augment local datasets with data
synthesized using differentially private means of latent data representations
communicated by a trusted server. Such augmentation ameliorates effects of data
heterogeneity across the clients without compromising privacy. Our experiments
on deep image classification tasks demonstrate that FedDPMS outperforms
competing state-of-the-art FL methods specifically designed for heterogeneous
data settings.
|
2207.13506 | Shan Wang | Shan Wang, Yanhao Zhang, Ankit Vora, Akhil Perincherry, and Hongdong
Li | Satellite Image Based Cross-view Localization for Autonomous Vehicle | Accepted by ICRA2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Existing spatial localization techniques for autonomous vehicles mostly use a
pre-built 3D-HD map, often constructed using a survey-grade 3D mapping vehicle,
which is not only expensive but also laborious. This paper shows that by using
an off-the-shelf high-definition satellite image as a ready-to-use map, we are
able to achieve cross-view vehicle localization up to a satisfactory accuracy,
providing a cheaper and more practical way for localization. While the
utilization of satellite imagery for cross-view localization is an established
concept, the conventional methodology focuses primarily on image retrieval.
This paper introduces a novel approach to cross-view localization that departs
from the conventional image retrieval method. Specifically, our method develops
(1) a Geometric-align Feature Extractor (GaFE) that leverages measured 3D
points to bridge the geometric gap between ground and overhead views, (2) a
Pose Aware Branch (PAB) adopting a triplet loss to encourage pose-aware feature
extraction, and (3) a Recursive Pose Refine Branch (RPRB) using the
Levenberg-Marquardt (LM) algorithm to align the initial pose towards the true
vehicle pose iteratively. Our method is validated on KITTI and Ford Multi-AV
Seasonal datasets as ground view and Google Maps as the satellite view. The
results demonstrate the superiority of our method in cross-view localization
with median spatial and angular errors within $1$ meter and $1^\circ$,
respectively.
| [
{
"version": "v1",
"created": "Wed, 27 Jul 2022 13:16:39 GMT"
},
{
"version": "v2",
"created": "Thu, 1 Sep 2022 02:48:56 GMT"
},
{
"version": "v3",
"created": "Thu, 20 Apr 2023 15:03:13 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Wang",
"Shan",
""
],
[
"Zhang",
"Yanhao",
""
],
[
"Vora",
"Ankit",
""
],
[
"Perincherry",
"Akhil",
""
],
[
"Li",
"Hongdong",
""
]
] | TITLE: Satellite Image Based Cross-view Localization for Autonomous Vehicle
ABSTRACT: Existing spatial localization techniques for autonomous vehicles mostly use a
pre-built 3D-HD map, often constructed using a survey-grade 3D mapping vehicle,
which is not only expensive but also laborious. This paper shows that by using
an off-the-shelf high-definition satellite image as a ready-to-use map, we are
able to achieve cross-view vehicle localization up to a satisfactory accuracy,
providing a cheaper and more practical way for localization. While the
utilization of satellite imagery for cross-view localization is an established
concept, the conventional methodology focuses primarily on image retrieval.
This paper introduces a novel approach to cross-view localization that departs
from the conventional image retrieval method. Specifically, our method develops
(1) a Geometric-align Feature Extractor (GaFE) that leverages measured 3D
points to bridge the geometric gap between ground and overhead views, (2) a
Pose Aware Branch (PAB) adopting a triplet loss to encourage pose-aware feature
extraction, and (3) a Recursive Pose Refine Branch (RPRB) using the
Levenberg-Marquardt (LM) algorithm to align the initial pose towards the true
vehicle pose iteratively. Our method is validated on KITTI and Ford Multi-AV
Seasonal datasets as ground view and Google Maps as the satellite view. The
results demonstrate the superiority of our method in cross-view localization
with median spatial and angular errors within $1$ meter and $1^\circ$,
respectively.
|
2209.05661 | Hugo Soulat | William I.Walker, Hugo Soulat, Changmin Yu, Maneesh Sahani | Unsupervised representation learning with recognition-parametrised
probabilistic models | null | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We introduce a new approach to probabilistic unsupervised learning based on
the recognition-parametrised model (RPM): a normalised semi-parametric
hypothesis class for joint distributions over observed and latent variables.
Under the key assumption that observations are conditionally independent given
latents, the RPM combines parametric prior and observation-conditioned latent
distributions with non-parametric observation marginals. This approach leads to
a flexible learnt recognition model capturing latent dependence between
observations, without the need for an explicit, parametric generative model.
The RPM admits exact maximum-likelihood learning for discrete latents, even for
powerful neural-network-based recognition. We develop effective approximations
applicable in the continuous-latent case. Experiments demonstrate the
effectiveness of the RPM on high-dimensional data, learning image
classification from weak indirect supervision; direct image-level latent
Dirichlet allocation; and recognition-parametrised Gaussian process factor
analysis (RP-GPFA) applied to multi-factorial spatiotemporal datasets. The RPM
provides a powerful framework to discover meaningful latent structure
underlying observational data, a function critical to both animal and
artificial intelligence.
| [
{
"version": "v1",
"created": "Tue, 13 Sep 2022 00:33:21 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 13:01:11 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Walker",
"William I.",
""
],
[
"Soulat",
"Hugo",
""
],
[
"Yu",
"Changmin",
""
],
[
"Sahani",
"Maneesh",
""
]
] | TITLE: Unsupervised representation learning with recognition-parametrised
probabilistic models
ABSTRACT: We introduce a new approach to probabilistic unsupervised learning based on
the recognition-parametrised model (RPM): a normalised semi-parametric
hypothesis class for joint distributions over observed and latent variables.
Under the key assumption that observations are conditionally independent given
latents, the RPM combines parametric prior and observation-conditioned latent
distributions with non-parametric observation marginals. This approach leads to
a flexible learnt recognition model capturing latent dependence between
observations, without the need for an explicit, parametric generative model.
The RPM admits exact maximum-likelihood learning for discrete latents, even for
powerful neural-network-based recognition. We develop effective approximations
applicable in the continuous-latent case. Experiments demonstrate the
effectiveness of the RPM on high-dimensional data, learning image
classification from weak indirect supervision; direct image-level latent
Dirichlet allocation; and recognition-parametrised Gaussian process factor
analysis (RP-GPFA) applied to multi-factorial spatiotemporal datasets. The RPM
provides a powerful framework to discover meaningful latent structure
underlying observational data, a function critical to both animal and
artificial intelligence.
|
2209.15171 | Zhuoran Qiao | Zhuoran Qiao, Weili Nie, Arash Vahdat, Thomas F. Miller III, Anima
Anandkumar | State-specific protein-ligand complex structure prediction with a
multi-scale deep generative model | 19 pages, 5 figures, 1 table & Supplementary Information (18 pages, 2
figures, 7 tables, 12 algorithms); supersedes an earlier version
arXiv:2209.15171v1 presented at the NeurIPS 2022 MLSB workshop as a
contributed talk | null | null | null | q-bio.QM cs.LG q-bio.BM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The binding complexes formed by proteins and small molecule ligands are
ubiquitous and critical to life. Despite recent advancements in protein
structure prediction, existing algorithms are so far unable to systematically
predict the binding ligand structures along with their regulatory effects on
protein folding. To address this discrepancy, we present NeuralPLexer, a
computational approach that can directly predict protein-ligand complex
structures solely using protein sequence and ligand molecular graph inputs.
NeuralPLexer adopts a deep generative model to sample the 3D structures of the
binding complex and their conformational changes at an atomistic resolution.
The model is based on a diffusion process that incorporates essential
biophysical constraints and a multi-scale geometric deep learning system to
iteratively sample residue-level contact maps and all heavy-atom coordinates in
a hierarchical manner. NeuralPLexer achieves state-of-the-art performance
compared to all existing methods on benchmarks for both protein-ligand blind
docking and flexible binding site structure recovery. Moreover, owing to its
specificity in sampling both ligand-free-state and ligand-bound-state
ensembles, NeuralPLexer consistently outperforms AlphaFold2 in terms of global
protein structure accuracy on both representative structure pairs with large
conformational changes (average TM-score=0.93) and recently determined
ligand-binding proteins (average TM-score=0.89). Case studies reveal that the
predicted conformational variations are consistent with structure determination
experiments for important targets, including human KRAS$^\textrm{G12C}$,
ketol-acid reductoisomerase, and purine GPCRs. Our study suggests that a
data-driven approach can capture the structural cooperativity between proteins
and small molecules, showing promise in accelerating the design of enzymes,
drug molecules, and beyond.
| [
{
"version": "v1",
"created": "Fri, 30 Sep 2022 01:46:38 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 19:40:00 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Qiao",
"Zhuoran",
""
],
[
"Nie",
"Weili",
""
],
[
"Vahdat",
"Arash",
""
],
[
"Miller",
"Thomas F.",
"III"
],
[
"Anandkumar",
"Anima",
""
]
] | TITLE: State-specific protein-ligand complex structure prediction with a
multi-scale deep generative model
ABSTRACT: The binding complexes formed by proteins and small molecule ligands are
ubiquitous and critical to life. Despite recent advancements in protein
structure prediction, existing algorithms are so far unable to systematically
predict the binding ligand structures along with their regulatory effects on
protein folding. To address this discrepancy, we present NeuralPLexer, a
computational approach that can directly predict protein-ligand complex
structures solely using protein sequence and ligand molecular graph inputs.
NeuralPLexer adopts a deep generative model to sample the 3D structures of the
binding complex and their conformational changes at an atomistic resolution.
The model is based on a diffusion process that incorporates essential
biophysical constraints and a multi-scale geometric deep learning system to
iteratively sample residue-level contact maps and all heavy-atom coordinates in
a hierarchical manner. NeuralPLexer achieves state-of-the-art performance
compared to all existing methods on benchmarks for both protein-ligand blind
docking and flexible binding site structure recovery. Moreover, owing to its
specificity in sampling both ligand-free-state and ligand-bound-state
ensembles, NeuralPLexer consistently outperforms AlphaFold2 in terms of global
protein structure accuracy on both representative structure pairs with large
conformational changes (average TM-score=0.93) and recently determined
ligand-binding proteins (average TM-score=0.89). Case studies reveal that the
predicted conformational variations are consistent with structure determination
experiments for important targets, including human KRAS$^\textrm{G12C}$,
ketol-acid reductoisomerase, and purine GPCRs. Our study suggests that a
data-driven approach can capture the structural cooperativity between proteins
and small molecules, showing promise in accelerating the design of enzymes,
drug molecules, and beyond.
|
2210.06688 | Weijun Tan | Weijun Tan, Qi Yao, Jingfeng Liu | Overlooked Video Classification in Weakly Supervised Video Anomaly
Detection | arXiv admin note: text overlap with arXiv:2101.10030 by other authors | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Current weakly supervised video anomaly detection algorithms mostly use
multiple instance learning (MIL) or their varieties. Almost all recent
approaches focus on how to select the correct snippets for training to improve
the performance. They overlook or do not realize the power of video
classification in boosting the performance of anomaly detection. In this paper,
we study explicitly the power of video classification supervision using a BERT
or LSTM. With this BERT or LSTM, CNN features of all snippets of a video can be
aggregated into a single feature which can be used for video classification.
This simple yet powerful video classification supervision, combined into the
MIL framework, brings extraordinary performance improvement on all three major
video anomaly detection datasets. Particularly it improves the mean average
precision (mAP) on the XD-Violence from SOTA 78.84\% to new 82.10\%. The source
code is available at
https://github.com/wjtan99/BERT_Anomaly_Video_Classification.
| [
{
"version": "v1",
"created": "Thu, 13 Oct 2022 03:00:22 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 22:23:33 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Tan",
"Weijun",
""
],
[
"Yao",
"Qi",
""
],
[
"Liu",
"Jingfeng",
""
]
] | TITLE: Overlooked Video Classification in Weakly Supervised Video Anomaly
Detection
ABSTRACT: Current weakly supervised video anomaly detection algorithms mostly use
multiple instance learning (MIL) or their varieties. Almost all recent
approaches focus on how to select the correct snippets for training to improve
the performance. They overlook or do not realize the power of video
classification in boosting the performance of anomaly detection. In this paper,
we study explicitly the power of video classification supervision using a BERT
or LSTM. With this BERT or LSTM, CNN features of all snippets of a video can be
aggregated into a single feature which can be used for video classification.
This simple yet powerful video classification supervision, combined into the
MIL framework, brings extraordinary performance improvement on all three major
video anomaly detection datasets. Particularly it improves the mean average
precision (mAP) on the XD-Violence from SOTA 78.84\% to new 82.10\%. The source
code is available at
https://github.com/wjtan99/BERT_Anomaly_Video_Classification.
|
2210.15827 | Ye Lin Tun | Ye Lin Tun, Chu Myaet Thwal, Yu Min Park, Seong-Bae Park, Choong Seon
Hong | Federated Learning with Intermediate Representation Regularization | IEEE BigComp 2023 | null | 10.1109/BigComp57234.2023.00017 | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In contrast to centralized model training that involves data collection,
federated learning (FL) enables remote clients to collaboratively train a model
without exposing their private data. However, model performance usually
degrades in FL due to the heterogeneous data generated by clients of diverse
characteristics. One promising strategy to maintain good performance is by
limiting the local training from drifting far away from the global model.
Previous studies accomplish this by regularizing the distance between the
representations learned by the local and global models. However, they only
consider representations from the early layers of a model or the layer
preceding the output layer. In this study, we introduce FedIntR, which provides
a more fine-grained regularization by integrating the representations of
intermediate layers into the local training process. Specifically, FedIntR
computes a regularization term that encourages the closeness between the
intermediate layer representations of the local and global models.
Additionally, FedIntR automatically determines the contribution of each layer's
representation to the regularization term based on the similarity between local
and global representations. We conduct extensive experiments on various
datasets to show that FedIntR can achieve equivalent or higher performance
compared to the state-of-the-art approaches. Our code is available at
https://github.com/YLTun/FedIntR.
| [
{
"version": "v1",
"created": "Fri, 28 Oct 2022 01:43:55 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 07:53:52 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Tun",
"Ye Lin",
""
],
[
"Thwal",
"Chu Myaet",
""
],
[
"Park",
"Yu Min",
""
],
[
"Park",
"Seong-Bae",
""
],
[
"Hong",
"Choong Seon",
""
]
] | TITLE: Federated Learning with Intermediate Representation Regularization
ABSTRACT: In contrast to centralized model training that involves data collection,
federated learning (FL) enables remote clients to collaboratively train a model
without exposing their private data. However, model performance usually
degrades in FL due to the heterogeneous data generated by clients of diverse
characteristics. One promising strategy to maintain good performance is by
limiting the local training from drifting far away from the global model.
Previous studies accomplish this by regularizing the distance between the
representations learned by the local and global models. However, they only
consider representations from the early layers of a model or the layer
preceding the output layer. In this study, we introduce FedIntR, which provides
a more fine-grained regularization by integrating the representations of
intermediate layers into the local training process. Specifically, FedIntR
computes a regularization term that encourages the closeness between the
intermediate layer representations of the local and global models.
Additionally, FedIntR automatically determines the contribution of each layer's
representation to the regularization term based on the similarity between local
and global representations. We conduct extensive experiments on various
datasets to show that FedIntR can achieve equivalent or higher performance
compared to the state-of-the-art approaches. Our code is available at
https://github.com/YLTun/FedIntR.
|
2212.00793 | Nithin Gopalakrishnan Nair | Nithin Gopalakrishnan Nair, Wele Gedara Chaminda Bandara and Vishal M.
Patel | Unite and Conquer: Plug & Play Multi-Modal Synthesis using Diffusion
Models | Accepted at CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Generating photos satisfying multiple constraints find broad utility in the
content creation industry. A key hurdle to accomplishing this task is the need
for paired data consisting of all modalities (i.e., constraints) and their
corresponding output. Moreover, existing methods need retraining using paired
data across all modalities to introduce a new condition. This paper proposes a
solution to this problem based on denoising diffusion probabilistic models
(DDPMs). Our motivation for choosing diffusion models over other generative
models comes from the flexible internal structure of diffusion models. Since
each sampling step in the DDPM follows a Gaussian distribution, we show that
there exists a closed-form solution for generating an image given various
constraints. Our method can unite multiple diffusion models trained on multiple
sub-tasks and conquer the combined task through our proposed sampling strategy.
We also introduce a novel reliability parameter that allows using different
off-the-shelf diffusion models trained across various datasets during sampling
time alone to guide it to the desired outcome satisfying multiple constraints.
We perform experiments on various standard multimodal tasks to demonstrate the
effectiveness of our approach. More details can be found in
https://nithin-gk.github.io/projectpages/Multidiff/index.html
| [
{
"version": "v1",
"created": "Thu, 1 Dec 2022 18:59:55 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 15:03:54 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Nair",
"Nithin Gopalakrishnan",
""
],
[
"Bandara",
"Wele Gedara Chaminda",
""
],
[
"Patel",
"Vishal M.",
""
]
] | TITLE: Unite and Conquer: Plug & Play Multi-Modal Synthesis using Diffusion
Models
ABSTRACT: Generating photos satisfying multiple constraints find broad utility in the
content creation industry. A key hurdle to accomplishing this task is the need
for paired data consisting of all modalities (i.e., constraints) and their
corresponding output. Moreover, existing methods need retraining using paired
data across all modalities to introduce a new condition. This paper proposes a
solution to this problem based on denoising diffusion probabilistic models
(DDPMs). Our motivation for choosing diffusion models over other generative
models comes from the flexible internal structure of diffusion models. Since
each sampling step in the DDPM follows a Gaussian distribution, we show that
there exists a closed-form solution for generating an image given various
constraints. Our method can unite multiple diffusion models trained on multiple
sub-tasks and conquer the combined task through our proposed sampling strategy.
We also introduce a novel reliability parameter that allows using different
off-the-shelf diffusion models trained across various datasets during sampling
time alone to guide it to the desired outcome satisfying multiple constraints.
We perform experiments on various standard multimodal tasks to demonstrate the
effectiveness of our approach. More details can be found in
https://nithin-gk.github.io/projectpages/Multidiff/index.html
|
2212.02397 | Mayank Baranwal | Anandsingh Chauhan, Mayank Baranwal, Ansuma Basumatary | PowRL: A Reinforcement Learning Framework for Robust Management of Power
Networks | Accepted at the 37th AAAI Conference on Artificial Intelligence | null | null | null | cs.LG cs.AI cs.SY eess.SY stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Power grids, across the world, play an important societal and economical role
by providing uninterrupted, reliable and transient-free power to several
industries, businesses and household consumers. With the advent of renewable
power resources and EVs resulting into uncertain generation and highly dynamic
load demands, it has become ever so important to ensure robust operation of
power networks through suitable management of transient stability issues and
localize the events of blackouts. In the light of ever increasing stress on the
modern grid infrastructure and the grid operators, this paper presents a
reinforcement learning (RL) framework, PowRL, to mitigate the effects of
unexpected network events, as well as reliably maintain electricity everywhere
on the network at all times. The PowRL leverages a novel heuristic for overload
management, along with the RL-guided decision making on optimal topology
selection to ensure that the grid is operated safely and reliably (with no
overloads). PowRL is benchmarked on a variety of competition datasets hosted by
the L2RPN (Learning to Run a Power Network). Even with its reduced action
space, PowRL tops the leaderboard in the L2RPN NeurIPS 2020 challenge
(Robustness track) at an aggregate level, while also being the top performing
agent in the L2RPN WCCI 2020 challenge. Moreover, detailed analysis depicts
state-of-the-art performances by the PowRL agent in some of the test scenarios.
| [
{
"version": "v1",
"created": "Mon, 5 Dec 2022 16:22:12 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 04:44:07 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Chauhan",
"Anandsingh",
""
],
[
"Baranwal",
"Mayank",
""
],
[
"Basumatary",
"Ansuma",
""
]
] | TITLE: PowRL: A Reinforcement Learning Framework for Robust Management of Power
Networks
ABSTRACT: Power grids, across the world, play an important societal and economical role
by providing uninterrupted, reliable and transient-free power to several
industries, businesses and household consumers. With the advent of renewable
power resources and EVs resulting into uncertain generation and highly dynamic
load demands, it has become ever so important to ensure robust operation of
power networks through suitable management of transient stability issues and
localize the events of blackouts. In the light of ever increasing stress on the
modern grid infrastructure and the grid operators, this paper presents a
reinforcement learning (RL) framework, PowRL, to mitigate the effects of
unexpected network events, as well as reliably maintain electricity everywhere
on the network at all times. The PowRL leverages a novel heuristic for overload
management, along with the RL-guided decision making on optimal topology
selection to ensure that the grid is operated safely and reliably (with no
overloads). PowRL is benchmarked on a variety of competition datasets hosted by
the L2RPN (Learning to Run a Power Network). Even with its reduced action
space, PowRL tops the leaderboard in the L2RPN NeurIPS 2020 challenge
(Robustness track) at an aggregate level, while also being the top performing
agent in the L2RPN WCCI 2020 challenge. Moreover, detailed analysis depicts
state-of-the-art performances by the PowRL agent in some of the test scenarios.
|
2212.05989 | Zhaonan Wang | Renhe Jiang, Zhaonan Wang, Jiawei Yong, Puneet Jeph, Quanjun Chen,
Yasumasa Kobayashi, Xuan Song, Toyotaro Suzumura, Shintaro Fukushima | MegaCRN: Meta-Graph Convolutional Recurrent Network for Spatio-Temporal
Modeling | Rejected by AIJ. We withdraw for now and shall further work on the
manuscript | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spatio-temporal modeling as a canonical task of multivariate time series
forecasting has been a significant research topic in AI community. To address
the underlying heterogeneity and non-stationarity implied in the graph streams,
in this study, we propose Spatio-Temporal Meta-Graph Learning as a novel Graph
Structure Learning mechanism on spatio-temporal data. Specifically, we
implement this idea into Meta-Graph Convolutional Recurrent Network (MegaCRN)
by plugging the Meta-Graph Learner powered by a Meta-Node Bank into GCRN
encoder-decoder. We conduct a comprehensive evaluation on two benchmark
datasets (METR-LA and PEMS-BAY) and a large-scale spatio-temporal dataset that
contains a variaty of non-stationary phenomena. Our model outperformed the
state-of-the-arts to a large degree on all three datasets (over 27% MAE and 34%
RMSE). Besides, through a series of qualitative evaluations, we demonstrate
that our model can explicitly disentangle locations and time slots with
different patterns and be robustly adaptive to different anomalous situations.
Codes and datasets are available at https://github.com/deepkashiwa20/MegaCRN.
| [
{
"version": "v1",
"created": "Mon, 12 Dec 2022 15:46:25 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 02:41:27 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Jiang",
"Renhe",
""
],
[
"Wang",
"Zhaonan",
""
],
[
"Yong",
"Jiawei",
""
],
[
"Jeph",
"Puneet",
""
],
[
"Chen",
"Quanjun",
""
],
[
"Kobayashi",
"Yasumasa",
""
],
[
"Song",
"Xuan",
""
],
[
"Suzumura",
"Toyotaro",
""
],
[
"Fukushima",
"Shintaro",
""
]
] | TITLE: MegaCRN: Meta-Graph Convolutional Recurrent Network for Spatio-Temporal
Modeling
ABSTRACT: Spatio-temporal modeling as a canonical task of multivariate time series
forecasting has been a significant research topic in AI community. To address
the underlying heterogeneity and non-stationarity implied in the graph streams,
in this study, we propose Spatio-Temporal Meta-Graph Learning as a novel Graph
Structure Learning mechanism on spatio-temporal data. Specifically, we
implement this idea into Meta-Graph Convolutional Recurrent Network (MegaCRN)
by plugging the Meta-Graph Learner powered by a Meta-Node Bank into GCRN
encoder-decoder. We conduct a comprehensive evaluation on two benchmark
datasets (METR-LA and PEMS-BAY) and a large-scale spatio-temporal dataset that
contains a variaty of non-stationary phenomena. Our model outperformed the
state-of-the-arts to a large degree on all three datasets (over 27% MAE and 34%
RMSE). Besides, through a series of qualitative evaluations, we demonstrate
that our model can explicitly disentangle locations and time slots with
different patterns and be robustly adaptive to different anomalous situations.
Codes and datasets are available at https://github.com/deepkashiwa20/MegaCRN.
|
2212.06313 | Seyed Jalaleddin Mousavirad | Seyed Jalaleddin Mousavirad, Lu\'is A Alexandre | Metaheuristic-based Energy-aware Image Compression for Mobile App
Development | This paper is submitted to the related journal | null | null | null | cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The JPEG standard is widely used in different image processing applications.
One of the main components of the JPEG standard is the quantisation table (QT)
since it plays a vital role in the image properties such as image quality and
file size. In recent years, several efforts based on population-based
metaheuristic (PBMH) algorithms have been performed to find the proper QT(s)
for a specific image, although they do not take into consideration the user
opinion in advance. Take an android developer as an example, who prefers a
small-size image, while the optimisation process results in a high-quality
image, leading to a huge file size. Another pitfall of the current works is a
lack of comprehensive coverage, meaning that the QT(s) can not provide all
possible combinations of file size and quality. Therefore, this paper aims to
propose three distinct contributions. First, to include the user opinion in the
compression process, the file size of the output image can be controlled by a
user in advance. To this end, we propose a novel objective function for
population-based JPEG image compression. Second, to tackle the lack of
comprehensive coverage, we suggest a novel representation. Our proposed
representation can not only provide more comprehensive coverage but also find
the proper value for the quality factor for a specific image without any
background knowledge. Both changes in representation and objective function are
independent of the search strategies and can be used with any type of
population-based metaheuristic (PBMH) algorithm. Therefore, as the third
contribution, we also provide a comprehensive benchmark on 22 state-of-the-art
and recently-introduced PBMH algorithms. Our extensive experiments on different
benchmark images and in terms of different criteria show that our novel
formulation for JPEG image compression can work effectively.
| [
{
"version": "v1",
"created": "Tue, 13 Dec 2022 01:39:47 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 14:38:48 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Mousavirad",
"Seyed Jalaleddin",
""
],
[
"Alexandre",
"Luís A",
""
]
] | TITLE: Metaheuristic-based Energy-aware Image Compression for Mobile App
Development
ABSTRACT: The JPEG standard is widely used in different image processing applications.
One of the main components of the JPEG standard is the quantisation table (QT)
since it plays a vital role in the image properties such as image quality and
file size. In recent years, several efforts based on population-based
metaheuristic (PBMH) algorithms have been performed to find the proper QT(s)
for a specific image, although they do not take into consideration the user
opinion in advance. Take an android developer as an example, who prefers a
small-size image, while the optimisation process results in a high-quality
image, leading to a huge file size. Another pitfall of the current works is a
lack of comprehensive coverage, meaning that the QT(s) can not provide all
possible combinations of file size and quality. Therefore, this paper aims to
propose three distinct contributions. First, to include the user opinion in the
compression process, the file size of the output image can be controlled by a
user in advance. To this end, we propose a novel objective function for
population-based JPEG image compression. Second, to tackle the lack of
comprehensive coverage, we suggest a novel representation. Our proposed
representation can not only provide more comprehensive coverage but also find
the proper value for the quality factor for a specific image without any
background knowledge. Both changes in representation and objective function are
independent of the search strategies and can be used with any type of
population-based metaheuristic (PBMH) algorithm. Therefore, as the third
contribution, we also provide a comprehensive benchmark on 22 state-of-the-art
and recently-introduced PBMH algorithms. Our extensive experiments on different
benchmark images and in terms of different criteria show that our novel
formulation for JPEG image compression can work effectively.
|
2301.12721 | Jianheng Tang | Jianheng Tang, Weiqi Zhang, Jiajin Li, Kangfei Zhao, Fugee Tsung, Jia
Li | Robust Attributed Graph Alignment via Joint Structure Learning and
Optimal Transport | ICDE 2023. Code and data are released at
https://github.com/squareRoot3/SLOTAlign | null | null | null | cs.DB cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph alignment, which aims at identifying corresponding entities across
multiple networks, has been widely applied in various domains. As the graphs to
be aligned are usually constructed from different sources, the inconsistency
issues of structures and features between two graphs are ubiquitous in
real-world applications. Most existing methods follow the
``embed-then-cross-compare'' paradigm, which computes node embeddings in each
graph and then processes node correspondences based on cross-graph embedding
comparison. However, we find these methods are unstable and sub-optimal when
structure or feature inconsistency appears. To this end, we propose SLOTAlign,
an unsupervised graph alignment framework that jointly performs Structure
Learning and Optimal Transport Alignment. We convert graph alignment to an
optimal transport problem between two intra-graph matrices without the
requirement of cross-graph comparison. We further incorporate multi-view
structure learning to enhance graph representation power and reduce the effect
of structure and feature inconsistency inherited across graphs. Moreover, an
alternating scheme based algorithm has been developed to address the joint
optimization problem in SLOTAlign, and the provable convergence result is also
established. Finally, we conduct extensive experiments on six unsupervised
graph alignment datasets and the DBP15K knowledge graph (KG) alignment
benchmark dataset. The proposed SLOTAlign shows superior performance and
strongest robustness over seven unsupervised graph alignment methods and five
specialized KG alignment methods.
| [
{
"version": "v1",
"created": "Mon, 30 Jan 2023 08:41:36 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 15:04:48 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Tang",
"Jianheng",
""
],
[
"Zhang",
"Weiqi",
""
],
[
"Li",
"Jiajin",
""
],
[
"Zhao",
"Kangfei",
""
],
[
"Tsung",
"Fugee",
""
],
[
"Li",
"Jia",
""
]
] | TITLE: Robust Attributed Graph Alignment via Joint Structure Learning and
Optimal Transport
ABSTRACT: Graph alignment, which aims at identifying corresponding entities across
multiple networks, has been widely applied in various domains. As the graphs to
be aligned are usually constructed from different sources, the inconsistency
issues of structures and features between two graphs are ubiquitous in
real-world applications. Most existing methods follow the
``embed-then-cross-compare'' paradigm, which computes node embeddings in each
graph and then processes node correspondences based on cross-graph embedding
comparison. However, we find these methods are unstable and sub-optimal when
structure or feature inconsistency appears. To this end, we propose SLOTAlign,
an unsupervised graph alignment framework that jointly performs Structure
Learning and Optimal Transport Alignment. We convert graph alignment to an
optimal transport problem between two intra-graph matrices without the
requirement of cross-graph comparison. We further incorporate multi-view
structure learning to enhance graph representation power and reduce the effect
of structure and feature inconsistency inherited across graphs. Moreover, an
alternating scheme based algorithm has been developed to address the joint
optimization problem in SLOTAlign, and the provable convergence result is also
established. Finally, we conduct extensive experiments on six unsupervised
graph alignment datasets and the DBP15K knowledge graph (KG) alignment
benchmark dataset. The proposed SLOTAlign shows superior performance and
strongest robustness over seven unsupervised graph alignment methods and five
specialized KG alignment methods.
|
2301.13459 | Zhiyuan Li | Zhiyuan Li, Anca Ralescu | Learning Generalized Hybrid Proximity Representation for Image
Recognition | The paper has been accepted by the IEEE ICTAI 2022 | null | 10.1109/ICTAI56018.2022.00138 | null | cs.CV cs.LG math.PR stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, deep metric learning techniques received attention, as the learned
distance representations are useful to capture the similarity relationship
among samples and further improve the performance of various of supervised or
unsupervised learning tasks. We propose a novel supervised metric learning
method that can learn the distance metrics in both geometric and probabilistic
space for image recognition. In contrast to the previous metric learning
methods which usually focus on learning the distance metrics in Euclidean
space, our proposed method is able to learn better distance representation in a
hybrid approach. To achieve this, we proposed a Generalized Hybrid Metric Loss
(GHM-Loss) to learn the general hybrid proximity features from the image data
by controlling the trade-off between geometric proximity and probabilistic
proximity. To evaluate the effectiveness of our method, we first provide
theoretical derivations and proofs of the proposed loss function, then we
perform extensive experiments on two public datasets to show the advantage of
our method compared to other state-of-the-art metric learning methods.
| [
{
"version": "v1",
"created": "Tue, 31 Jan 2023 07:49:25 GMT"
},
{
"version": "v2",
"created": "Sun, 16 Apr 2023 03:22:41 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Li",
"Zhiyuan",
""
],
[
"Ralescu",
"Anca",
""
]
] | TITLE: Learning Generalized Hybrid Proximity Representation for Image
Recognition
ABSTRACT: Recently, deep metric learning techniques received attention, as the learned
distance representations are useful to capture the similarity relationship
among samples and further improve the performance of various of supervised or
unsupervised learning tasks. We propose a novel supervised metric learning
method that can learn the distance metrics in both geometric and probabilistic
space for image recognition. In contrast to the previous metric learning
methods which usually focus on learning the distance metrics in Euclidean
space, our proposed method is able to learn better distance representation in a
hybrid approach. To achieve this, we proposed a Generalized Hybrid Metric Loss
(GHM-Loss) to learn the general hybrid proximity features from the image data
by controlling the trade-off between geometric proximity and probabilistic
proximity. To evaluate the effectiveness of our method, we first provide
theoretical derivations and proofs of the proposed loss function, then we
perform extensive experiments on two public datasets to show the advantage of
our method compared to other state-of-the-art metric learning methods.
|
2302.10866 | Michael Poli | Michael Poli, Stefano Massaroli, Eric Nguyen, Daniel Y. Fu, Tri Dao,
Stephen Baccus, Yoshua Bengio, Stefano Ermon, Christopher R\'e | Hyena Hierarchy: Towards Larger Convolutional Language Models | Additional details | null | null | null | cs.LG cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in deep learning have relied heavily on the use of large
Transformers due to their ability to learn at scale. However, the core building
block of Transformers, the attention operator, exhibits quadratic cost in
sequence length, limiting the amount of context accessible. Existing
subquadratic methods based on low-rank and sparse approximations need to be
combined with dense attention layers to match Transformers, indicating a gap in
capability. In this work, we propose Hyena, a subquadratic drop-in replacement
for attention constructed by interleaving implicitly parametrized long
convolutions and data-controlled gating. In recall and reasoning tasks on
sequences of thousands to hundreds of thousands of tokens, Hyena improves
accuracy by more than 50 points over operators relying on state-spaces and
other implicit and explicit methods, matching attention-based models. We set a
new state-of-the-art for dense-attention-free architectures on language
modeling in standard datasets (WikiText103 and The Pile), reaching Transformer
quality with a 20% reduction in training compute required at sequence length
2K. Hyena operators are twice as fast as highly optimized attention at sequence
length 8K, and 100x faster at sequence length 64K.
| [
{
"version": "v1",
"created": "Tue, 21 Feb 2023 18:29:25 GMT"
},
{
"version": "v2",
"created": "Mon, 6 Mar 2023 01:26:15 GMT"
},
{
"version": "v3",
"created": "Wed, 19 Apr 2023 20:08:39 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Poli",
"Michael",
""
],
[
"Massaroli",
"Stefano",
""
],
[
"Nguyen",
"Eric",
""
],
[
"Fu",
"Daniel Y.",
""
],
[
"Dao",
"Tri",
""
],
[
"Baccus",
"Stephen",
""
],
[
"Bengio",
"Yoshua",
""
],
[
"Ermon",
"Stefano",
""
],
[
"Ré",
"Christopher",
""
]
] | TITLE: Hyena Hierarchy: Towards Larger Convolutional Language Models
ABSTRACT: Recent advances in deep learning have relied heavily on the use of large
Transformers due to their ability to learn at scale. However, the core building
block of Transformers, the attention operator, exhibits quadratic cost in
sequence length, limiting the amount of context accessible. Existing
subquadratic methods based on low-rank and sparse approximations need to be
combined with dense attention layers to match Transformers, indicating a gap in
capability. In this work, we propose Hyena, a subquadratic drop-in replacement
for attention constructed by interleaving implicitly parametrized long
convolutions and data-controlled gating. In recall and reasoning tasks on
sequences of thousands to hundreds of thousands of tokens, Hyena improves
accuracy by more than 50 points over operators relying on state-spaces and
other implicit and explicit methods, matching attention-based models. We set a
new state-of-the-art for dense-attention-free architectures on language
modeling in standard datasets (WikiText103 and The Pile), reaching Transformer
quality with a 20% reduction in training compute required at sequence length
2K. Hyena operators are twice as fast as highly optimized attention at sequence
length 8K, and 100x faster at sequence length 64K.
|
2302.13140 | Xiao Hu | Xiao Hu and Qichen Wang | Computing the Difference of Conjunctive Queries Efficiently | null | null | null | null | cs.DB | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We investigate how to efficiently compute the difference result of two (or
multiple) conjunctive queries, which is the last operator in relational algebra
to be unraveled. The standard approach in practical database systems is to
materialize the results for every input query as a separate set, and then
compute the difference of two (or multiple) sets. This approach is bottlenecked
by the complexity of evaluating every input query individually, which could be
very expensive, particularly when there are only a few results in the
difference. In this paper, we introduce a new approach by exploiting the
structural property of input queries and rewriting the original query by
pushing the difference operator down as much as possible. We show that for a
large class of difference queries, this approach can lead to a linear-time
algorithm, in terms of the input size and (final) output size, i.e., the number
of query results that survive from the difference operator. We complete this
result by showing the hardness of computing the remaining difference queries in
linear time. Although a linear-time algorithm is hard to achieve in general, we
also provide some heuristics that can provably improve the standard approach.
At last, we compare our approach with standard SQL engines over graph and
benchmark datasets. The experiment results demonstrate order-of-magnitude
speedups achieved by our approach over the vanilla SQL.
| [
{
"version": "v1",
"created": "Sat, 25 Feb 2023 18:42:54 GMT"
},
{
"version": "v2",
"created": "Thu, 9 Mar 2023 00:30:42 GMT"
},
{
"version": "v3",
"created": "Thu, 20 Apr 2023 12:24:23 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Hu",
"Xiao",
""
],
[
"Wang",
"Qichen",
""
]
] | TITLE: Computing the Difference of Conjunctive Queries Efficiently
ABSTRACT: We investigate how to efficiently compute the difference result of two (or
multiple) conjunctive queries, which is the last operator in relational algebra
to be unraveled. The standard approach in practical database systems is to
materialize the results for every input query as a separate set, and then
compute the difference of two (or multiple) sets. This approach is bottlenecked
by the complexity of evaluating every input query individually, which could be
very expensive, particularly when there are only a few results in the
difference. In this paper, we introduce a new approach by exploiting the
structural property of input queries and rewriting the original query by
pushing the difference operator down as much as possible. We show that for a
large class of difference queries, this approach can lead to a linear-time
algorithm, in terms of the input size and (final) output size, i.e., the number
of query results that survive from the difference operator. We complete this
result by showing the hardness of computing the remaining difference queries in
linear time. Although a linear-time algorithm is hard to achieve in general, we
also provide some heuristics that can provably improve the standard approach.
At last, we compare our approach with standard SQL engines over graph and
benchmark datasets. The experiment results demonstrate order-of-magnitude
speedups achieved by our approach over the vanilla SQL.
|
2302.14250 | Chaohui Yu | Chaohui Yu, Qiang Zhou, Jingliang Li, Jianlong Yuan, Zhibin Wang, Fan
Wang | Foundation Model Drives Weakly Incremental Learning for Semantic
Segmentation | CVPR 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Modern incremental learning for semantic segmentation methods usually learn
new categories based on dense annotations. Although achieve promising results,
pixel-by-pixel labeling is costly and time-consuming. Weakly incremental
learning for semantic segmentation (WILSS) is a novel and attractive task,
which aims at learning to segment new classes from cheap and widely available
image-level labels. Despite the comparable results, the image-level labels can
not provide details to locate each segment, which limits the performance of
WILSS. This inspires us to think how to improve and effectively utilize the
supervision of new classes given image-level labels while avoiding forgetting
old ones. In this work, we propose a novel and data-efficient framework for
WILSS, named FMWISS. Specifically, we propose pre-training based
co-segmentation to distill the knowledge of complementary foundation models for
generating dense pseudo labels. We further optimize the noisy pseudo masks with
a teacher-student architecture, where a plug-in teacher is optimized with a
proposed dense contrastive loss. Moreover, we introduce memory-based copy-paste
augmentation to improve the catastrophic forgetting problem of old classes.
Extensive experiments on Pascal VOC and COCO datasets demonstrate the superior
performance of our framework, e.g., FMWISS achieves 70.7% and 73.3% in the 15-5
VOC setting, outperforming the state-of-the-art method by 3.4% and 6.1%,
respectively.
| [
{
"version": "v1",
"created": "Tue, 28 Feb 2023 02:21:42 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 08:12:44 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Yu",
"Chaohui",
""
],
[
"Zhou",
"Qiang",
""
],
[
"Li",
"Jingliang",
""
],
[
"Yuan",
"Jianlong",
""
],
[
"Wang",
"Zhibin",
""
],
[
"Wang",
"Fan",
""
]
] | TITLE: Foundation Model Drives Weakly Incremental Learning for Semantic
Segmentation
ABSTRACT: Modern incremental learning for semantic segmentation methods usually learn
new categories based on dense annotations. Although achieve promising results,
pixel-by-pixel labeling is costly and time-consuming. Weakly incremental
learning for semantic segmentation (WILSS) is a novel and attractive task,
which aims at learning to segment new classes from cheap and widely available
image-level labels. Despite the comparable results, the image-level labels can
not provide details to locate each segment, which limits the performance of
WILSS. This inspires us to think how to improve and effectively utilize the
supervision of new classes given image-level labels while avoiding forgetting
old ones. In this work, we propose a novel and data-efficient framework for
WILSS, named FMWISS. Specifically, we propose pre-training based
co-segmentation to distill the knowledge of complementary foundation models for
generating dense pseudo labels. We further optimize the noisy pseudo masks with
a teacher-student architecture, where a plug-in teacher is optimized with a
proposed dense contrastive loss. Moreover, we introduce memory-based copy-paste
augmentation to improve the catastrophic forgetting problem of old classes.
Extensive experiments on Pascal VOC and COCO datasets demonstrate the superior
performance of our framework, e.g., FMWISS achieves 70.7% and 73.3% in the 15-5
VOC setting, outperforming the state-of-the-art method by 3.4% and 6.1%,
respectively.
|
2303.00304 | Obin Kwon | Obin Kwon, Jeongho Park, Songhwai Oh | Renderable Neural Radiance Map for Visual Navigation | Preprint version. CVPR 2023 accepted, highlight paper. Project page:
https://rllab-snu.github.io/projects/RNR-Map/ | null | null | null | cs.CV cs.RO | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We propose a novel type of map for visual navigation, a renderable neural
radiance map (RNR-Map), which is designed to contain the overall visual
information of a 3D environment. The RNR-Map has a grid form and consists of
latent codes at each pixel. These latent codes are embedded from image
observations, and can be converted to the neural radiance field which enables
image rendering given a camera pose. The recorded latent codes implicitly
contain visual information about the environment, which makes the RNR-Map
visually descriptive. This visual information in RNR-Map can be a useful
guideline for visual localization and navigation. We develop localization and
navigation frameworks that can effectively utilize the RNR-Map. We evaluate the
proposed frameworks on camera tracking, visual localization, and image-goal
navigation. Experimental results show that the RNR-Map-based localization
framework can find the target location based on a single query image with fast
speed and competitive accuracy compared to other baselines. Also, this
localization framework is robust to environmental changes, and even finds the
most visually similar places when a query image from a different environment is
given. The proposed navigation framework outperforms the existing image-goal
navigation methods in difficult scenarios, under odometry and actuation noises.
The navigation framework shows 65.7% success rate in curved scenarios of the
NRNS dataset, which is an improvement of 18.6% over the current
state-of-the-art. Project page: https://rllab-snu.github.io/projects/RNR-Map/
| [
{
"version": "v1",
"created": "Wed, 1 Mar 2023 08:00:46 GMT"
},
{
"version": "v2",
"created": "Fri, 3 Mar 2023 11:12:20 GMT"
},
{
"version": "v3",
"created": "Thu, 23 Mar 2023 05:59:24 GMT"
},
{
"version": "v4",
"created": "Thu, 20 Apr 2023 01:50:55 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Kwon",
"Obin",
""
],
[
"Park",
"Jeongho",
""
],
[
"Oh",
"Songhwai",
""
]
] | TITLE: Renderable Neural Radiance Map for Visual Navigation
ABSTRACT: We propose a novel type of map for visual navigation, a renderable neural
radiance map (RNR-Map), which is designed to contain the overall visual
information of a 3D environment. The RNR-Map has a grid form and consists of
latent codes at each pixel. These latent codes are embedded from image
observations, and can be converted to the neural radiance field which enables
image rendering given a camera pose. The recorded latent codes implicitly
contain visual information about the environment, which makes the RNR-Map
visually descriptive. This visual information in RNR-Map can be a useful
guideline for visual localization and navigation. We develop localization and
navigation frameworks that can effectively utilize the RNR-Map. We evaluate the
proposed frameworks on camera tracking, visual localization, and image-goal
navigation. Experimental results show that the RNR-Map-based localization
framework can find the target location based on a single query image with fast
speed and competitive accuracy compared to other baselines. Also, this
localization framework is robust to environmental changes, and even finds the
most visually similar places when a query image from a different environment is
given. The proposed navigation framework outperforms the existing image-goal
navigation methods in difficult scenarios, under odometry and actuation noises.
The navigation framework shows 65.7% success rate in curved scenarios of the
NRNS dataset, which is an improvement of 18.6% over the current
state-of-the-art. Project page: https://rllab-snu.github.io/projects/RNR-Map/
|
2303.13769 | Feng Xue | Wenteng Liang, Feng Xue, Yihao Liu, Guofeng Zhong, Anlong Ming | Unknown Sniffer for Object Detection: Don't Turn a Blind Eye to Unknown
Objects | CVPR 2023 camera-ready; Code: https://github.com/Went-Liang/UnSniffer
Project: https://xuefengbupt.github.io/project_page/unsniffer_cvpr23.html
Demo: https://www.bilibili.com/video/BV1xM4y1z7Hv Supplymentary:
https://xuefengbupt.github.io/project_page/pdf/supplementary_cvpr2023.pdf | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The recently proposed open-world object and open-set detection have achieved
a breakthrough in finding never-seen-before objects and distinguishing them
from known ones. However, their studies on knowledge transfer from known
classes to unknown ones are not deep enough, resulting in the scanty capability
for detecting unknowns hidden in the background. In this paper, we propose the
unknown sniffer (UnSniffer) to find both unknown and known objects. Firstly,
the generalized object confidence (GOC) score is introduced, which only uses
known samples for supervision and avoids improper suppression of unknowns in
the background. Significantly, such confidence score learned from known objects
can be generalized to unknown ones. Additionally, we propose a negative energy
suppression loss to further suppress the non-object samples in the background.
Next, the best box of each unknown is hard to obtain during inference due to
lacking their semantic information in training. To solve this issue, we
introduce a graph-based determination scheme to replace hand-designed
non-maximum suppression (NMS) post-processing. Finally, we present the Unknown
Object Detection Benchmark, the first publicly benchmark that encompasses
precision evaluation for unknown detection to our knowledge. Experiments show
that our method is far better than the existing state-of-the-art methods.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 02:59:44 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 13:28:25 GMT"
},
{
"version": "v3",
"created": "Thu, 20 Apr 2023 01:19:34 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Liang",
"Wenteng",
""
],
[
"Xue",
"Feng",
""
],
[
"Liu",
"Yihao",
""
],
[
"Zhong",
"Guofeng",
""
],
[
"Ming",
"Anlong",
""
]
] | TITLE: Unknown Sniffer for Object Detection: Don't Turn a Blind Eye to Unknown
Objects
ABSTRACT: The recently proposed open-world object and open-set detection have achieved
a breakthrough in finding never-seen-before objects and distinguishing them
from known ones. However, their studies on knowledge transfer from known
classes to unknown ones are not deep enough, resulting in the scanty capability
for detecting unknowns hidden in the background. In this paper, we propose the
unknown sniffer (UnSniffer) to find both unknown and known objects. Firstly,
the generalized object confidence (GOC) score is introduced, which only uses
known samples for supervision and avoids improper suppression of unknowns in
the background. Significantly, such confidence score learned from known objects
can be generalized to unknown ones. Additionally, we propose a negative energy
suppression loss to further suppress the non-object samples in the background.
Next, the best box of each unknown is hard to obtain during inference due to
lacking their semantic information in training. To solve this issue, we
introduce a graph-based determination scheme to replace hand-designed
non-maximum suppression (NMS) post-processing. Finally, we present the Unknown
Object Detection Benchmark, the first publicly benchmark that encompasses
precision evaluation for unknown detection to our knowledge. Experiments show
that our method is far better than the existing state-of-the-art methods.
|
2303.13842 | Kailun Yang | Hao Shi, Yu Li, Kailun Yang, Jiaming Zhang, Kunyu Peng, Alina
Roitberg, Yaozu Ye, Huajian Ni, Kaiwei Wang, Rainer Stiefelhagen | FishDreamer: Towards Fisheye Semantic Completion via Unified Image
Outpainting and Segmentation | Accepted to CVPR OmniCV 2023. Code and datasets will be available at
https://github.com/MasterHow/FishDreamer | null | null | null | cs.CV cs.RO eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper raises the new task of Fisheye Semantic Completion (FSC), where
dense texture, structure, and semantics of a fisheye image are inferred even
beyond the sensor field-of-view (FoV). Fisheye cameras have larger FoV than
ordinary pinhole cameras, yet its unique special imaging model naturally leads
to a blind area at the edge of the image plane. This is suboptimal for
safety-critical applications since important perception tasks, such as semantic
segmentation, become very challenging within the blind zone. Previous works
considered the out-FoV outpainting and in-FoV segmentation separately. However,
we observe that these two tasks are actually closely coupled. To jointly
estimate the tightly intertwined complete fisheye image and scene semantics, we
introduce the new FishDreamer which relies on successful ViTs enhanced with a
novel Polar-aware Cross Attention module (PCA) to leverage dense context and
guide semantically-consistent content generation while considering different
polar distributions. In addition to the contribution of the novel task and
architecture, we also derive Cityscapes-BF and KITTI360-BF datasets to
facilitate training and evaluation of this new track. Our experiments
demonstrate that the proposed FishDreamer outperforms methods solving each task
in isolation and surpasses alternative approaches on the Fisheye Semantic
Completion. Code and datasets are publicly available at
https://github.com/MasterHow/FishDreamer.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 07:34:25 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 12:27:36 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Shi",
"Hao",
""
],
[
"Li",
"Yu",
""
],
[
"Yang",
"Kailun",
""
],
[
"Zhang",
"Jiaming",
""
],
[
"Peng",
"Kunyu",
""
],
[
"Roitberg",
"Alina",
""
],
[
"Ye",
"Yaozu",
""
],
[
"Ni",
"Huajian",
""
],
[
"Wang",
"Kaiwei",
""
],
[
"Stiefelhagen",
"Rainer",
""
]
] | TITLE: FishDreamer: Towards Fisheye Semantic Completion via Unified Image
Outpainting and Segmentation
ABSTRACT: This paper raises the new task of Fisheye Semantic Completion (FSC), where
dense texture, structure, and semantics of a fisheye image are inferred even
beyond the sensor field-of-view (FoV). Fisheye cameras have larger FoV than
ordinary pinhole cameras, yet its unique special imaging model naturally leads
to a blind area at the edge of the image plane. This is suboptimal for
safety-critical applications since important perception tasks, such as semantic
segmentation, become very challenging within the blind zone. Previous works
considered the out-FoV outpainting and in-FoV segmentation separately. However,
we observe that these two tasks are actually closely coupled. To jointly
estimate the tightly intertwined complete fisheye image and scene semantics, we
introduce the new FishDreamer which relies on successful ViTs enhanced with a
novel Polar-aware Cross Attention module (PCA) to leverage dense context and
guide semantically-consistent content generation while considering different
polar distributions. In addition to the contribution of the novel task and
architecture, we also derive Cityscapes-BF and KITTI360-BF datasets to
facilitate training and evaluation of this new track. Our experiments
demonstrate that the proposed FishDreamer outperforms methods solving each task
in isolation and surpasses alternative approaches on the Fisheye Semantic
Completion. Code and datasets are publicly available at
https://github.com/MasterHow/FishDreamer.
|
2304.00086 | Ajit Desai | Ajit Desai | Machine Learning for Economics Research: When What and How? | null | null | null | null | econ.GN cs.AI cs.LG q-fin.EC stat.AP | http://creativecommons.org/licenses/by/4.0/ | This article provides a curated review of selected papers published in
prominent economics journals that use machine learning (ML) tools for research
and policy analysis. The review focuses on three key questions: (1) when ML is
used in economics, (2) what ML models are commonly preferred, and (3) how they
are used for economic applications. The review highlights that ML is
particularly used to process nontraditional and unstructured data, capture
strong nonlinearity, and improve prediction accuracy. Deep learning models are
suitable for nontraditional data, whereas ensemble learning models are
preferred for traditional datasets. While traditional econometric models may
suffice for analyzing low-complexity data, the increasing complexity of
economic data due to rapid digitalization and the growing literature suggests
that ML is becoming an essential addition to the econometrician's toolbox.
| [
{
"version": "v1",
"created": "Fri, 31 Mar 2023 19:21:56 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 16:30:01 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Desai",
"Ajit",
""
]
] | TITLE: Machine Learning for Economics Research: When What and How?
ABSTRACT: This article provides a curated review of selected papers published in
prominent economics journals that use machine learning (ML) tools for research
and policy analysis. The review focuses on three key questions: (1) when ML is
used in economics, (2) what ML models are commonly preferred, and (3) how they
are used for economic applications. The review highlights that ML is
particularly used to process nontraditional and unstructured data, capture
strong nonlinearity, and improve prediction accuracy. Deep learning models are
suitable for nontraditional data, whereas ensemble learning models are
preferred for traditional datasets. While traditional econometric models may
suffice for analyzing low-complexity data, the increasing complexity of
economic data due to rapid digitalization and the growing literature suggests
that ML is becoming an essential addition to the econometrician's toolbox.
|
2304.02122 | Joe Yue-Hei Ng | Joe Yue-Hei Ng, Kevin McCloskey, Jian Cui, Vincent R. Meijer, Erica
Brand, Aaron Sarna, Nita Goyal, Christopher Van Arsdale, Scott Geraedts | OpenContrails: Benchmarking Contrail Detection on GOES-16 ABI | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Contrails (condensation trails) are line-shaped ice clouds caused by aircraft
and are likely the largest contributor of aviation-induced climate change.
Contrail avoidance is potentially an inexpensive way to significantly reduce
the climate impact of aviation. An automated contrail detection system is an
essential tool to develop and evaluate contrail avoidance systems. In this
paper, we present a human-labeled dataset named OpenContrails to train and
evaluate contrail detection models based on GOES-16 Advanced Baseline Imager
(ABI) data. We propose and evaluate a contrail detection model that
incorporates temporal context for improved detection accuracy. The human
labeled dataset and the contrail detection outputs are publicly available on
Google Cloud Storage at gs://goes_contrails_dataset.
| [
{
"version": "v1",
"created": "Tue, 4 Apr 2023 21:03:46 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 06:00:41 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Ng",
"Joe Yue-Hei",
""
],
[
"McCloskey",
"Kevin",
""
],
[
"Cui",
"Jian",
""
],
[
"Meijer",
"Vincent R.",
""
],
[
"Brand",
"Erica",
""
],
[
"Sarna",
"Aaron",
""
],
[
"Goyal",
"Nita",
""
],
[
"Van Arsdale",
"Christopher",
""
],
[
"Geraedts",
"Scott",
""
]
] | TITLE: OpenContrails: Benchmarking Contrail Detection on GOES-16 ABI
ABSTRACT: Contrails (condensation trails) are line-shaped ice clouds caused by aircraft
and are likely the largest contributor of aviation-induced climate change.
Contrail avoidance is potentially an inexpensive way to significantly reduce
the climate impact of aviation. An automated contrail detection system is an
essential tool to develop and evaluate contrail avoidance systems. In this
paper, we present a human-labeled dataset named OpenContrails to train and
evaluate contrail detection models based on GOES-16 Advanced Baseline Imager
(ABI) data. We propose and evaluate a contrail detection model that
incorporates temporal context for improved detection accuracy. The human
labeled dataset and the contrail detection outputs are publicly available on
Google Cloud Storage at gs://goes_contrails_dataset.
|
2304.04087 | Tanveer Ahmed Belal | Tanveer Ahmed Belal, G. M. Shahariar, Md. Hasanul Kabir | Interpretable Multi Labeled Bengali Toxic Comments Classification using
Deep Learning | null | 2023 International Conference on Electrical, Computer and
Communication Engineering (ECCE) | 10.1109/ECCE57851.2023.10101588 | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This paper presents a deep learning-based pipeline for categorizing Bengali
toxic comments, in which at first a binary classification model is used to
determine whether a comment is toxic or not, and then a multi-label classifier
is employed to determine which toxicity type the comment belongs to. For this
purpose, we have prepared a manually labeled dataset consisting of 16,073
instances among which 8,488 are Toxic and any toxic comment may correspond to
one or more of the six toxic categories - vulgar, hate, religious, threat,
troll, and insult simultaneously. Long Short Term Memory (LSTM) with BERT
Embedding achieved 89.42% accuracy for the binary classification task while as
a multi-label classifier, a combination of Convolutional Neural Network and
Bi-directional Long Short Term Memory (CNN-BiLSTM) with attention mechanism
achieved 78.92% accuracy and 0.86 as weighted F1-score. To explain the
predictions and interpret the word feature importance during classification by
the proposed models, we utilized Local Interpretable Model-Agnostic
Explanations (LIME) framework. We have made our dataset public and can be
accessed at -
https://github.com/deepu099cse/Multi-Labeled-Bengali-Toxic-Comments-Classification
| [
{
"version": "v1",
"created": "Sat, 8 Apr 2023 19:28:26 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Belal",
"Tanveer Ahmed",
""
],
[
"Shahariar",
"G. M.",
""
],
[
"Kabir",
"Md. Hasanul",
""
]
] | TITLE: Interpretable Multi Labeled Bengali Toxic Comments Classification using
Deep Learning
ABSTRACT: This paper presents a deep learning-based pipeline for categorizing Bengali
toxic comments, in which at first a binary classification model is used to
determine whether a comment is toxic or not, and then a multi-label classifier
is employed to determine which toxicity type the comment belongs to. For this
purpose, we have prepared a manually labeled dataset consisting of 16,073
instances among which 8,488 are Toxic and any toxic comment may correspond to
one or more of the six toxic categories - vulgar, hate, religious, threat,
troll, and insult simultaneously. Long Short Term Memory (LSTM) with BERT
Embedding achieved 89.42% accuracy for the binary classification task while as
a multi-label classifier, a combination of Convolutional Neural Network and
Bi-directional Long Short Term Memory (CNN-BiLSTM) with attention mechanism
achieved 78.92% accuracy and 0.86 as weighted F1-score. To explain the
predictions and interpret the word feature importance during classification by
the proposed models, we utilized Local Interpretable Model-Agnostic
Explanations (LIME) framework. We have made our dataset public and can be
accessed at -
https://github.com/deepu099cse/Multi-Labeled-Bengali-Toxic-Comments-Classification
|
2304.08756 | Yang Liu | Yang Liu, Shen Yan, Yuge Zhang, Kan Ren, Quanlu Zhang, Zebin Ren, Deng
Cai, Mi Zhang | AutoTaskFormer: Searching Vision Transformers for Multi-task Learning | 15 pages | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Vision Transformers have shown great performance in single tasks such as
classification and segmentation. However, real-world problems are not isolated,
which calls for vision transformers that can perform multiple tasks
concurrently. Existing multi-task vision transformers are handcrafted and
heavily rely on human expertise. In this work, we propose a novel one-shot
neural architecture search framework, dubbed AutoTaskFormer (Automated
Multi-Task Vision TransFormer), to automate this process. AutoTaskFormer not
only identifies the weights to share across multiple tasks automatically, but
also provides thousands of well-trained vision transformers with a wide range
of parameters (e.g., number of heads and network depth) for deployment under
various resource constraints. Experiments on both small-scale (2-task
Cityscapes and 3-task NYUv2) and large-scale (16-task Taskonomy) datasets show
that AutoTaskFormer outperforms state-of-the-art handcrafted vision
transformers in multi-task learning. The entire code and models will be
open-sourced.
| [
{
"version": "v1",
"created": "Tue, 18 Apr 2023 06:30:20 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 02:27:04 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Liu",
"Yang",
""
],
[
"Yan",
"Shen",
""
],
[
"Zhang",
"Yuge",
""
],
[
"Ren",
"Kan",
""
],
[
"Zhang",
"Quanlu",
""
],
[
"Ren",
"Zebin",
""
],
[
"Cai",
"Deng",
""
],
[
"Zhang",
"Mi",
""
]
] | TITLE: AutoTaskFormer: Searching Vision Transformers for Multi-task Learning
ABSTRACT: Vision Transformers have shown great performance in single tasks such as
classification and segmentation. However, real-world problems are not isolated,
which calls for vision transformers that can perform multiple tasks
concurrently. Existing multi-task vision transformers are handcrafted and
heavily rely on human expertise. In this work, we propose a novel one-shot
neural architecture search framework, dubbed AutoTaskFormer (Automated
Multi-Task Vision TransFormer), to automate this process. AutoTaskFormer not
only identifies the weights to share across multiple tasks automatically, but
also provides thousands of well-trained vision transformers with a wide range
of parameters (e.g., number of heads and network depth) for deployment under
various resource constraints. Experiments on both small-scale (2-task
Cityscapes and 3-task NYUv2) and large-scale (16-task Taskonomy) datasets show
that AutoTaskFormer outperforms state-of-the-art handcrafted vision
transformers in multi-task learning. The entire code and models will be
open-sourced.
|
2304.09616 | Josu Goikoetxea | J. Goikoetxea, M. Arantzeta, I. San Martin | Bridging Natural Language Processing and Psycholinguistics:
computationally grounded semantic similarity datasets for Basque and Spanish | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | We present a computationally-grounded word similarity dataset based on two
well-known Natural Language Processing resources; text corpora and knowledge
bases. This dataset aims to fulfil a gap in psycholinguistic research by
providing a variety of quantifications of semantic similarity in an extensive
set of noun pairs controlled by variables that play a significant role in
lexical processing. The dataset creation has consisted in three steps, 1)
computing four key psycholinguistic features for each noun; concreteness,
frequency, semantic and phonological neighbourhood density; 2) pairing nouns
across these four variables; 3) for each noun pair, assigning three types of
word similarity measurements, computed out of text, Wordnet and hybrid
embeddings. The present dataset includes noun pairs' information in Basque and
European Spanish, but further work intends to extend it to more languages.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 12:47:51 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 08:23:21 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Goikoetxea",
"J.",
""
],
[
"Arantzeta",
"M.",
""
],
[
"Martin",
"I. San",
""
]
] | TITLE: Bridging Natural Language Processing and Psycholinguistics:
computationally grounded semantic similarity datasets for Basque and Spanish
ABSTRACT: We present a computationally-grounded word similarity dataset based on two
well-known Natural Language Processing resources; text corpora and knowledge
bases. This dataset aims to fulfil a gap in psycholinguistic research by
providing a variety of quantifications of semantic similarity in an extensive
set of noun pairs controlled by variables that play a significant role in
lexical processing. The dataset creation has consisted in three steps, 1)
computing four key psycholinguistic features for each noun; concreteness,
frequency, semantic and phonological neighbourhood density; 2) pairing nouns
across these four variables; 3) for each noun pair, assigning three types of
word similarity measurements, computed out of text, Wordnet and hybrid
embeddings. The present dataset includes noun pairs' information in Basque and
European Spanish, but further work intends to extend it to more languages.
|
2304.09623 | Pranav Jeevan P | Chirag P, Mukta Wagle, Ravi Kant Gupta, Pranav Jeevan, Amit Sethi | CHATTY: Coupled Holistic Adversarial Transport Terms with Yield for
Unsupervised Domain Adaptation | 10 pages, 4 figures | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | We propose a new technique called CHATTY: Coupled Holistic Adversarial
Transport Terms with Yield for Unsupervised Domain Adaptation. Adversarial
training is commonly used for learning domain-invariant representations by
reversing the gradients from a domain discriminator head to train the feature
extractor layers of a neural network. We propose significant modifications to
the adversarial head, its training objective, and the classifier head. With the
aim of reducing class confusion, we introduce a sub-network which displaces the
classifier outputs of the source and target domain samples in a learnable
manner. We control this movement using a novel transport loss that spreads
class clusters away from each other and makes it easier for the classifier to
find the decision boundaries for both the source and target domains. The
results of adding this new loss to a careful selection of previously proposed
losses leads to improvement in UDA results compared to the previous
state-of-the-art methods on benchmark datasets. We show the importance of the
proposed loss term using ablation studies and visualization of the movement of
target domain sample in representation space.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 13:00:23 GMT"
},
{
"version": "v2",
"created": "Thu, 20 Apr 2023 16:39:43 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"P",
"Chirag",
""
],
[
"Wagle",
"Mukta",
""
],
[
"Gupta",
"Ravi Kant",
""
],
[
"Jeevan",
"Pranav",
""
],
[
"Sethi",
"Amit",
""
]
] | TITLE: CHATTY: Coupled Holistic Adversarial Transport Terms with Yield for
Unsupervised Domain Adaptation
ABSTRACT: We propose a new technique called CHATTY: Coupled Holistic Adversarial
Transport Terms with Yield for Unsupervised Domain Adaptation. Adversarial
training is commonly used for learning domain-invariant representations by
reversing the gradients from a domain discriminator head to train the feature
extractor layers of a neural network. We propose significant modifications to
the adversarial head, its training objective, and the classifier head. With the
aim of reducing class confusion, we introduce a sub-network which displaces the
classifier outputs of the source and target domain samples in a learnable
manner. We control this movement using a novel transport loss that spreads
class clusters away from each other and makes it easier for the classifier to
find the decision boundaries for both the source and target domains. The
results of adding this new loss to a careful selection of previously proposed
losses leads to improvement in UDA results compared to the previous
state-of-the-art methods on benchmark datasets. We show the importance of the
proposed loss term using ablation studies and visualization of the movement of
target domain sample in representation space.
|
2304.09859 | Daniel G. Krakowczyk | Daniel G. Krakowczyk, David R. Reich, Jakob Chwastek, Deborah N.
Jakobi, Paul Prasse, Assunta S\"uss, Oleksii Turuta, Pawe{\l} Kasprowski,
Lena A. J\"ager | pymovements: A Python Package for Eye Movement Data Processing | Preprint for ETRA '23: 2023 Symposium on Eye Tracking Research and
Applications | null | 10.1145/3588015.3590134 | null | cs.HC | http://creativecommons.org/licenses/by/4.0/ | We introduce pymovements: a Python package for analyzing eye-tracking data
that follows best practices in software development, including rigorous testing
and adherence to coding standards. The package provides functionality for key
processes along the entire preprocessing pipeline. This includes parsing of eye
tracker data files, transforming positional data into velocity data, detecting
gaze events like saccades and fixations, computing event properties like
saccade amplitude and fixational dispersion and visualizing data and results
with several types of plotting methods. Moreover, pymovements also provides an
easily accessible interface for downloading and processing publicly available
datasets. Additionally, we emphasize how rigorous testing in scientific
software packages is critical to the reproducibility and transparency of
research, enabling other researchers to verify and build upon previous
findings.
| [
{
"version": "v1",
"created": "Tue, 11 Apr 2023 18:39:37 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Krakowczyk",
"Daniel G.",
""
],
[
"Reich",
"David R.",
""
],
[
"Chwastek",
"Jakob",
""
],
[
"Jakobi",
"Deborah N.",
""
],
[
"Prasse",
"Paul",
""
],
[
"Süss",
"Assunta",
""
],
[
"Turuta",
"Oleksii",
""
],
[
"Kasprowski",
"Paweł",
""
],
[
"Jäger",
"Lena A.",
""
]
] | TITLE: pymovements: A Python Package for Eye Movement Data Processing
ABSTRACT: We introduce pymovements: a Python package for analyzing eye-tracking data
that follows best practices in software development, including rigorous testing
and adherence to coding standards. The package provides functionality for key
processes along the entire preprocessing pipeline. This includes parsing of eye
tracker data files, transforming positional data into velocity data, detecting
gaze events like saccades and fixations, computing event properties like
saccade amplitude and fixational dispersion and visualizing data and results
with several types of plotting methods. Moreover, pymovements also provides an
easily accessible interface for downloading and processing publicly available
datasets. Additionally, we emphasize how rigorous testing in scientific
software packages is critical to the reproducibility and transparency of
research, enabling other researchers to verify and build upon previous
findings.
|
2304.09867 | Susanna Raj | Susanna Raj, Sudha Jamthe, Yashaswini Viswanath, Suresh Lokiah | Introducing Construct Theory as a Standard Methodology for Inclusive AI
Models | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Construct theory in social psychology, developed by George Kelly are mental
constructs to predict and anticipate events. Constructs are how humans
interpret, curate, predict and validate data; information. AI today is biased
because it is trained with a narrow construct as defined by the training data
labels. Machine Learning algorithms for facial recognition discriminate against
darker skin colors and in the ground breaking research papers (Buolamwini, Joy
and Timnit Gebru. Gender Shades: Intersectional Accuracy Disparities in
Commercial Gender Classification. FAT (2018), the inclusion of phenotypic
labeling is proposed as a viable solution. In Construct theory, phenotype is
just one of the many subelements that make up the construct of a face. In this
paper, we present 15 main elements of the construct of face, with 50
subelements and tested Google Cloud Vision API and Microsoft Cognitive Services
API using FairFace dataset that currently has data for 7 races, genders and
ages, and we retested against FairFace Plus dataset curated by us. Our results
show exactly where they have gaps for inclusivity. Based on our experiment
results, we propose that validated, inclusive constructs become industry
standards for AI ML models going forward.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 01:01:19 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Raj",
"Susanna",
""
],
[
"Jamthe",
"Sudha",
""
],
[
"Viswanath",
"Yashaswini",
""
],
[
"Lokiah",
"Suresh",
""
]
] | TITLE: Introducing Construct Theory as a Standard Methodology for Inclusive AI
Models
ABSTRACT: Construct theory in social psychology, developed by George Kelly are mental
constructs to predict and anticipate events. Constructs are how humans
interpret, curate, predict and validate data; information. AI today is biased
because it is trained with a narrow construct as defined by the training data
labels. Machine Learning algorithms for facial recognition discriminate against
darker skin colors and in the ground breaking research papers (Buolamwini, Joy
and Timnit Gebru. Gender Shades: Intersectional Accuracy Disparities in
Commercial Gender Classification. FAT (2018), the inclusion of phenotypic
labeling is proposed as a viable solution. In Construct theory, phenotype is
just one of the many subelements that make up the construct of a face. In this
paper, we present 15 main elements of the construct of face, with 50
subelements and tested Google Cloud Vision API and Microsoft Cognitive Services
API using FairFace dataset that currently has data for 7 races, genders and
ages, and we retested against FairFace Plus dataset curated by us. Our results
show exactly where they have gaps for inclusivity. Based on our experiment
results, we propose that validated, inclusive constructs become industry
standards for AI ML models going forward.
|
2304.09869 | Chengpeng Hu | Chengpeng Hu, Jiyuan Pei, Jialin Liu, Xin Yao | Evolving Constrained Reinforcement Learning Policy | accepted by the 2023 International Joint Conference on Neural
Networks (IJCNN) | null | null | null | cs.NE cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Evolutionary algorithms have been used to evolve a population of actors to
generate diverse experiences for training reinforcement learning agents, which
helps to tackle the temporal credit assignment problem and improves the
exploration efficiency. However, when adapting this approach to address
constrained problems, balancing the trade-off between the reward and constraint
violation is hard. In this paper, we propose a novel evolutionary constrained
reinforcement learning (ECRL) algorithm, which adaptively balances the reward
and constraint violation with stochastic ranking, and at the same time,
restricts the policy's behaviour by maintaining a set of Lagrange relaxation
coefficients with a constraint buffer. Extensive experiments on robotic control
benchmarks show that our ECRL achieves outstanding performance compared to
state-of-the-art algorithms. Ablation analysis shows the benefits of
introducing stochastic ranking and constraint buffer.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 03:54:31 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Hu",
"Chengpeng",
""
],
[
"Pei",
"Jiyuan",
""
],
[
"Liu",
"Jialin",
""
],
[
"Yao",
"Xin",
""
]
] | TITLE: Evolving Constrained Reinforcement Learning Policy
ABSTRACT: Evolutionary algorithms have been used to evolve a population of actors to
generate diverse experiences for training reinforcement learning agents, which
helps to tackle the temporal credit assignment problem and improves the
exploration efficiency. However, when adapting this approach to address
constrained problems, balancing the trade-off between the reward and constraint
violation is hard. In this paper, we propose a novel evolutionary constrained
reinforcement learning (ECRL) algorithm, which adaptively balances the reward
and constraint violation with stochastic ranking, and at the same time,
restricts the policy's behaviour by maintaining a set of Lagrange relaxation
coefficients with a constraint buffer. Extensive experiments on robotic control
benchmarks show that our ECRL achieves outstanding performance compared to
state-of-the-art algorithms. Ablation analysis shows the benefits of
introducing stochastic ranking and constraint buffer.
|
2304.09874 | Prakash Chandra Chhipa | Muskaan Chopra, Prakash Chandra Chhipa, Gopal Mengi, Varun Gupta and
Marcus Liwicki | Domain Adaptable Self-supervised Representation Learning on Remote
Sensing Satellite Imagery | Accepted in International Joint Conference on Neural Networks (IJCNN)
2023. First three authors shares equal contribution! | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | This work presents a novel domain adaption paradigm for studying contrastive
self-supervised representation learning and knowledge transfer using remote
sensing satellite data. Major state-of-the-art remote sensing visual domain
efforts primarily focus on fully supervised learning approaches that rely
entirely on human annotations. On the other hand, human annotations in remote
sensing satellite imagery are always subject to limited quantity due to high
costs and domain expertise, making transfer learning a viable alternative. The
proposed approach investigates the knowledge transfer of selfsupervised
representations across the distinct source and target data distributions in
depth in the remote sensing data domain. In this arrangement, self-supervised
contrastive learning-based pretraining is performed on the source dataset, and
downstream tasks are performed on the target datasets in a round-robin fashion.
Experiments are conducted on three publicly available datasets, UC Merced
Landuse (UCMD), SIRI-WHU, and MLRSNet, for different downstream classification
tasks versus label efficiency. In self-supervised knowledge transfer, the
proposed approach achieves state-of-the-art performance with label efficiency
labels and outperforms a fully supervised setting. A more in-depth qualitative
examination reveals consistent evidence for explainable representation
learning. The source code and trained models are published on GitHub.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 14:32:36 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Chopra",
"Muskaan",
""
],
[
"Chhipa",
"Prakash Chandra",
""
],
[
"Mengi",
"Gopal",
""
],
[
"Gupta",
"Varun",
""
],
[
"Liwicki",
"Marcus",
""
]
] | TITLE: Domain Adaptable Self-supervised Representation Learning on Remote
Sensing Satellite Imagery
ABSTRACT: This work presents a novel domain adaption paradigm for studying contrastive
self-supervised representation learning and knowledge transfer using remote
sensing satellite data. Major state-of-the-art remote sensing visual domain
efforts primarily focus on fully supervised learning approaches that rely
entirely on human annotations. On the other hand, human annotations in remote
sensing satellite imagery are always subject to limited quantity due to high
costs and domain expertise, making transfer learning a viable alternative. The
proposed approach investigates the knowledge transfer of selfsupervised
representations across the distinct source and target data distributions in
depth in the remote sensing data domain. In this arrangement, self-supervised
contrastive learning-based pretraining is performed on the source dataset, and
downstream tasks are performed on the target datasets in a round-robin fashion.
Experiments are conducted on three publicly available datasets, UC Merced
Landuse (UCMD), SIRI-WHU, and MLRSNet, for different downstream classification
tasks versus label efficiency. In self-supervised knowledge transfer, the
proposed approach achieves state-of-the-art performance with label efficiency
labels and outperforms a fully supervised setting. A more in-depth qualitative
examination reveals consistent evidence for explainable representation
learning. The source code and trained models are published on GitHub.
|
2304.09913 | Kyungsu Kim | Sanghyun Jo, In-Jae Yu, Kyungsu Kim | MARS: Model-agnostic Biased Object Removal without Additional
Supervision for Weakly-Supervised Semantic Segmentation | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Weakly-supervised semantic segmentation aims to reduce labeling costs by
training semantic segmentation models using weak supervision, such as
image-level class labels. However, most approaches struggle to produce accurate
localization maps and suffer from false predictions in class-related
backgrounds (i.e., biased objects), such as detecting a railroad with the train
class. Recent methods that remove biased objects require additional supervision
for manually identifying biased objects for each problematic class and
collecting their datasets by reviewing predictions, limiting their
applicability to the real-world dataset with multiple labels and complex
relationships for biasing. Following the first observation that biased features
can be separated and eliminated by matching biased objects with backgrounds in
the same dataset, we propose a fully-automatic/model-agnostic biased removal
framework called MARS (Model-Agnostic biased object Removal without additional
Supervision), which utilizes semantically consistent features of an
unsupervised technique to eliminate biased objects in pseudo labels.
Surprisingly, we show that MARS achieves new state-of-the-art results on two
popular benchmarks, PASCAL VOC 2012 (val: 77.7%, test: 77.2%) and MS COCO 2014
(val: 49.4%), by consistently improving the performance of various WSSS models
by at least 30% without additional supervision.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 18:24:28 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Jo",
"Sanghyun",
""
],
[
"Yu",
"In-Jae",
""
],
[
"Kim",
"Kyungsu",
""
]
] | TITLE: MARS: Model-agnostic Biased Object Removal without Additional
Supervision for Weakly-Supervised Semantic Segmentation
ABSTRACT: Weakly-supervised semantic segmentation aims to reduce labeling costs by
training semantic segmentation models using weak supervision, such as
image-level class labels. However, most approaches struggle to produce accurate
localization maps and suffer from false predictions in class-related
backgrounds (i.e., biased objects), such as detecting a railroad with the train
class. Recent methods that remove biased objects require additional supervision
for manually identifying biased objects for each problematic class and
collecting their datasets by reviewing predictions, limiting their
applicability to the real-world dataset with multiple labels and complex
relationships for biasing. Following the first observation that biased features
can be separated and eliminated by matching biased objects with backgrounds in
the same dataset, we propose a fully-automatic/model-agnostic biased removal
framework called MARS (Model-Agnostic biased object Removal without additional
Supervision), which utilizes semantically consistent features of an
unsupervised technique to eliminate biased objects in pseudo labels.
Surprisingly, we show that MARS achieves new state-of-the-art results on two
popular benchmarks, PASCAL VOC 2012 (val: 77.7%, test: 77.2%) and MS COCO 2014
(val: 49.4%), by consistently improving the performance of various WSSS models
by at least 30% without additional supervision.
|
2304.09919 | Marcus Schwarting | Vesa Akerman and David Baines and Damien Daspit and Ulf Hermjakob and
Taeho Jang and Colin Leong and Michael Martin and Joel Mathew and Jonathan
Robie and Marcus Schwarting | The eBible Corpus: Data and Model Benchmarks for Bible Translation for
Low-Resource Languages | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Efficiently and accurately translating a corpus into a low-resource language
remains a challenge, regardless of the strategies employed, whether manual,
automated, or a combination of the two. Many Christian organizations are
dedicated to the task of translating the Holy Bible into languages that lack a
modern translation. Bible translation (BT) work is currently underway for over
3000 extremely low resource languages. We introduce the eBible corpus: a
dataset containing 1009 translations of portions of the Bible with data in 833
different languages across 75 language families. In addition to a BT
benchmarking dataset, we introduce model performance benchmarks built on the No
Language Left Behind (NLLB) neural machine translation (NMT) models. Finally,
we describe several problems specific to the domain of BT and consider how the
established data and model benchmarks might be used for future translation
efforts. For a BT task trained with NLLB, Austronesian and Trans-New Guinea
language families achieve 35.1 and 31.6 BLEU scores respectively, which spurs
future innovations for NMT for low-resource languages in Papua New Guinea.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 18:52:49 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Akerman",
"Vesa",
""
],
[
"Baines",
"David",
""
],
[
"Daspit",
"Damien",
""
],
[
"Hermjakob",
"Ulf",
""
],
[
"Jang",
"Taeho",
""
],
[
"Leong",
"Colin",
""
],
[
"Martin",
"Michael",
""
],
[
"Mathew",
"Joel",
""
],
[
"Robie",
"Jonathan",
""
],
[
"Schwarting",
"Marcus",
""
]
] | TITLE: The eBible Corpus: Data and Model Benchmarks for Bible Translation for
Low-Resource Languages
ABSTRACT: Efficiently and accurately translating a corpus into a low-resource language
remains a challenge, regardless of the strategies employed, whether manual,
automated, or a combination of the two. Many Christian organizations are
dedicated to the task of translating the Holy Bible into languages that lack a
modern translation. Bible translation (BT) work is currently underway for over
3000 extremely low resource languages. We introduce the eBible corpus: a
dataset containing 1009 translations of portions of the Bible with data in 833
different languages across 75 language families. In addition to a BT
benchmarking dataset, we introduce model performance benchmarks built on the No
Language Left Behind (NLLB) neural machine translation (NMT) models. Finally,
we describe several problems specific to the domain of BT and consider how the
established data and model benchmarks might be used for future translation
efforts. For a BT task trained with NLLB, Austronesian and Trans-New Guinea
language families achieve 35.1 and 31.6 BLEU scores respectively, which spurs
future innovations for NMT for low-resource languages in Papua New Guinea.
|
2304.09948 | Aishwarya Deep Shukla | Aishwarya Deep Shukla, Laksh Agarwal, Jie Mein (JM) Goh, Guodong
(Gordon) Gao, Ritu Agarwal | Catch Me If You Can: Identifying Fraudulent Physician Reviews with Large
Language Models Using Generative Pre-Trained Transformers | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The proliferation of fake reviews of doctors has potentially detrimental
consequences for patient well-being and has prompted concern among consumer
protection groups and regulatory bodies. Yet despite significant advancements
in the fields of machine learning and natural language processing, there
remains limited comprehension of the characteristics differentiating fraudulent
from authentic reviews. This study utilizes a novel pre-labeled dataset of
38048 physician reviews to establish the effectiveness of large language models
in classifying reviews. Specifically, we compare the performance of traditional
ML models, such as logistic regression and support vector machines, to
generative pre-trained transformer models. Furthermore, we use GPT4, the newest
model in the GPT family, to uncover the key dimensions along which fake and
genuine physician reviews differ. Our findings reveal significantly superior
performance of GPT-3 over traditional ML models in this context. Additionally,
our analysis suggests that GPT3 requires a smaller training sample than
traditional models, suggesting its appropriateness for tasks with scarce
training data. Moreover, the superiority of GPT3 performance increases in the
cold start context i.e., when there are no prior reviews of a doctor. Finally,
we employ GPT4 to reveal the crucial dimensions that distinguish fake physician
reviews. In sharp contrast to previous findings in the literature that were
obtained using simulated data, our findings from a real-world dataset show that
fake reviews are generally more clinically detailed, more reserved in
sentiment, and have better structure and grammar than authentic ones.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 19:59:26 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Shukla",
"Aishwarya Deep",
"",
"JM"
],
[
"Agarwal",
"Laksh",
"",
"JM"
],
[
"Mein",
"Jie",
"",
"JM"
],
[
"Goh",
"",
"",
"Gordon"
],
[
"Guodong",
"",
"",
"Gordon"
],
[
"Gao",
"",
""
],
[
"Agarwal",
"Ritu",
""
]
] | TITLE: Catch Me If You Can: Identifying Fraudulent Physician Reviews with Large
Language Models Using Generative Pre-Trained Transformers
ABSTRACT: The proliferation of fake reviews of doctors has potentially detrimental
consequences for patient well-being and has prompted concern among consumer
protection groups and regulatory bodies. Yet despite significant advancements
in the fields of machine learning and natural language processing, there
remains limited comprehension of the characteristics differentiating fraudulent
from authentic reviews. This study utilizes a novel pre-labeled dataset of
38048 physician reviews to establish the effectiveness of large language models
in classifying reviews. Specifically, we compare the performance of traditional
ML models, such as logistic regression and support vector machines, to
generative pre-trained transformer models. Furthermore, we use GPT4, the newest
model in the GPT family, to uncover the key dimensions along which fake and
genuine physician reviews differ. Our findings reveal significantly superior
performance of GPT-3 over traditional ML models in this context. Additionally,
our analysis suggests that GPT3 requires a smaller training sample than
traditional models, suggesting its appropriateness for tasks with scarce
training data. Moreover, the superiority of GPT3 performance increases in the
cold start context i.e., when there are no prior reviews of a doctor. Finally,
we employ GPT4 to reveal the crucial dimensions that distinguish fake physician
reviews. In sharp contrast to previous findings in the literature that were
obtained using simulated data, our findings from a real-world dataset show that
fake reviews are generally more clinically detailed, more reserved in
sentiment, and have better structure and grammar than authentic ones.
|
2304.09957 | Ekaterina Artemova | Ekaterina Artemova and Barbara Plank | Low-resource Bilingual Dialect Lexicon Induction with Large Language
Models | Accepted to NoDaLiDa 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Bilingual word lexicons are crucial tools for multilingual natural language
understanding and machine translation tasks, as they facilitate the mapping of
words in one language to their synonyms in another language. To achieve this,
numerous papers have explored bilingual lexicon induction (BLI) in
high-resource scenarios, using a typical pipeline consisting of two
unsupervised steps: bitext mining and word alignment, both of which rely on
pre-trained large language models~(LLMs).
In this paper, we present an analysis of the BLI pipeline for German and two
of its dialects, Bavarian and Alemannic. This setup poses several unique
challenges, including the scarcity of resources, the relatedness of the
languages, and the lack of standardization in the orthography of dialects. To
evaluate the BLI outputs, we analyze them with respect to word frequency and
pairwise edit distance. Additionally, we release two evaluation datasets
comprising 1,500 bilingual sentence pairs and 1,000 bilingual word pairs. They
were manually judged for their semantic similarity for each Bavarian-German and
Alemannic-German language pair.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 20:20:41 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Artemova",
"Ekaterina",
""
],
[
"Plank",
"Barbara",
""
]
] | TITLE: Low-resource Bilingual Dialect Lexicon Induction with Large Language
Models
ABSTRACT: Bilingual word lexicons are crucial tools for multilingual natural language
understanding and machine translation tasks, as they facilitate the mapping of
words in one language to their synonyms in another language. To achieve this,
numerous papers have explored bilingual lexicon induction (BLI) in
high-resource scenarios, using a typical pipeline consisting of two
unsupervised steps: bitext mining and word alignment, both of which rely on
pre-trained large language models~(LLMs).
In this paper, we present an analysis of the BLI pipeline for German and two
of its dialects, Bavarian and Alemannic. This setup poses several unique
challenges, including the scarcity of resources, the relatedness of the
languages, and the lack of standardization in the orthography of dialects. To
evaluate the BLI outputs, we analyze them with respect to word frequency and
pairwise edit distance. Additionally, we release two evaluation datasets
comprising 1,500 bilingual sentence pairs and 1,000 bilingual word pairs. They
were manually judged for their semantic similarity for each Bavarian-German and
Alemannic-German language pair.
|
2304.10004 | Jaime Sevilla | Ege Erdil and Jaime Sevilla | Power Law Trends in Speedrunning and Machine Learning | null | null | null | null | cs.LG cs.AI stat.AP | http://creativecommons.org/licenses/by/4.0/ | We find that improvements in speedrunning world records follow a power law
pattern. Using this observation, we answer an outstanding question from
previous work: How do we improve on the baseline of predicting no improvement
when forecasting speedrunning world records out to some time horizon, such as
one month? Using a random effects model, we improve on this baseline for
relative mean square error made on predicting out-of-sample world record
improvements as the comparison metric at a $p < 10^{-5}$ significance level.
The same set-up improves \textit{even} on the ex-post best exponential moving
average forecasts at a $p = 0.15$ significance level while having access to
substantially fewer data points. We demonstrate the effectiveness of this
approach by applying it to Machine Learning benchmarks and achieving forecasts
that exceed a baseline. Finally, we interpret the resulting model to suggest
that 1) ML benchmarks are far from saturation and 2) sudden large improvements
in Machine Learning are unlikely but cannot be ruled out.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 23:03:03 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Erdil",
"Ege",
""
],
[
"Sevilla",
"Jaime",
""
]
] | TITLE: Power Law Trends in Speedrunning and Machine Learning
ABSTRACT: We find that improvements in speedrunning world records follow a power law
pattern. Using this observation, we answer an outstanding question from
previous work: How do we improve on the baseline of predicting no improvement
when forecasting speedrunning world records out to some time horizon, such as
one month? Using a random effects model, we improve on this baseline for
relative mean square error made on predicting out-of-sample world record
improvements as the comparison metric at a $p < 10^{-5}$ significance level.
The same set-up improves \textit{even} on the ex-post best exponential moving
average forecasts at a $p = 0.15$ significance level while having access to
substantially fewer data points. We demonstrate the effectiveness of this
approach by applying it to Machine Learning benchmarks and achieving forecasts
that exceed a baseline. Finally, we interpret the resulting model to suggest
that 1) ML benchmarks are far from saturation and 2) sudden large improvements
in Machine Learning are unlikely but cannot be ruled out.
|
2304.10009 | Mir Sayed Shah Danish Asst. Prof. | Mir Sayed Shah Danish, Zahra Nazari, Tomonobu Senjyu | AI-coherent data-driven forecasting model for a combined cycle power
plant | 47 pages. 21 Figures, Tables 2, Journal of Energy Conversion and
Management | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | This study investigates the transformation of energy models to align with
machine learning requirements as a promising tool for optimizing the operation
of combined cycle power plants (CCPPs). By modeling energy production as a
function of environmental and control variables, this methodology offers an
innovative way to achieve energy-efficient power generation in the context of
the data-driven application. This study focuses on developing a thorough
AI-coherent modeling approach for CCPP optimization, preferring an
interdisciplinary perspective and coming up with a comprehensive, insightful
analysis. The proposed numerical model using Broyden Fletcher Goldfarb Shanno
(BFGS) algorithm enhances efficiency by simulating various operating scenarios
and adjusting optimal parameters, leading to a high yield power generation of
2.23% increase from 452 MW to 462.1 MW by optimizing the environmental factors.
This study deals with data-driven modeling based on historical data to make
predictions without prior knowledge of the system's parameter, demonstrating
several merits in identifying patterns that can be difficult for human analysts
to detect, high accuracy when trained on large datasets, and the potential to
improve over time with new data. The proposed modeling approach and methodology
can be expanded as a valuable tool for forecasting and decision-making in
complex energy systems.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 23:25:55 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Danish",
"Mir Sayed Shah",
""
],
[
"Nazari",
"Zahra",
""
],
[
"Senjyu",
"Tomonobu",
""
]
] | TITLE: AI-coherent data-driven forecasting model for a combined cycle power
plant
ABSTRACT: This study investigates the transformation of energy models to align with
machine learning requirements as a promising tool for optimizing the operation
of combined cycle power plants (CCPPs). By modeling energy production as a
function of environmental and control variables, this methodology offers an
innovative way to achieve energy-efficient power generation in the context of
the data-driven application. This study focuses on developing a thorough
AI-coherent modeling approach for CCPP optimization, preferring an
interdisciplinary perspective and coming up with a comprehensive, insightful
analysis. The proposed numerical model using Broyden Fletcher Goldfarb Shanno
(BFGS) algorithm enhances efficiency by simulating various operating scenarios
and adjusting optimal parameters, leading to a high yield power generation of
2.23% increase from 452 MW to 462.1 MW by optimizing the environmental factors.
This study deals with data-driven modeling based on historical data to make
predictions without prior knowledge of the system's parameter, demonstrating
several merits in identifying patterns that can be difficult for human analysts
to detect, high accuracy when trained on large datasets, and the potential to
improve over time with new data. The proposed modeling approach and methodology
can be expanded as a valuable tool for forecasting and decision-making in
complex energy systems.
|
2304.10029 | Bilel Tarchoun | Bilel Tarchoun, Anouar Ben Khalifa, Mohamed Ali Mahjoub, Nael
Abu-Ghazaleh, Ihsen Alouani | Jedi: Entropy-based Localization and Removal of Adversarial Patches | 9 pages, 11 figures. To appear in CVPR 2023 | null | null | null | cs.CR cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Real-world adversarial physical patches were shown to be successful in
compromising state-of-the-art models in a variety of computer vision
applications. Existing defenses that are based on either input gradient or
features analysis have been compromised by recent GAN-based attacks that
generate naturalistic patches. In this paper, we propose Jedi, a new defense
against adversarial patches that is resilient to realistic patch attacks. Jedi
tackles the patch localization problem from an information theory perspective;
leverages two new ideas: (1) it improves the identification of potential patch
regions using entropy analysis: we show that the entropy of adversarial patches
is high, even in naturalistic patches; and (2) it improves the localization of
adversarial patches, using an autoencoder that is able to complete patch
regions from high entropy kernels. Jedi achieves high-precision adversarial
patch localization, which we show is critical to successfully repair the
images. Since Jedi relies on an input entropy analysis, it is model-agnostic,
and can be applied on pre-trained off-the-shelf models without changes to the
training or inference of the protected models. Jedi detects on average 90% of
adversarial patches across different benchmarks and recovers up to 94% of
successful patch attacks (Compared to 75% and 65% for LGS and Jujutsu,
respectively).
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 00:56:59 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Tarchoun",
"Bilel",
""
],
[
"Khalifa",
"Anouar Ben",
""
],
[
"Mahjoub",
"Mohamed Ali",
""
],
[
"Abu-Ghazaleh",
"Nael",
""
],
[
"Alouani",
"Ihsen",
""
]
] | TITLE: Jedi: Entropy-based Localization and Removal of Adversarial Patches
ABSTRACT: Real-world adversarial physical patches were shown to be successful in
compromising state-of-the-art models in a variety of computer vision
applications. Existing defenses that are based on either input gradient or
features analysis have been compromised by recent GAN-based attacks that
generate naturalistic patches. In this paper, we propose Jedi, a new defense
against adversarial patches that is resilient to realistic patch attacks. Jedi
tackles the patch localization problem from an information theory perspective;
leverages two new ideas: (1) it improves the identification of potential patch
regions using entropy analysis: we show that the entropy of adversarial patches
is high, even in naturalistic patches; and (2) it improves the localization of
adversarial patches, using an autoencoder that is able to complete patch
regions from high entropy kernels. Jedi achieves high-precision adversarial
patch localization, which we show is critical to successfully repair the
images. Since Jedi relies on an input entropy analysis, it is model-agnostic,
and can be applied on pre-trained off-the-shelf models without changes to the
training or inference of the protected models. Jedi detects on average 90% of
adversarial patches across different benchmarks and recovers up to 94% of
successful patch attacks (Compared to 75% and 65% for LGS and Jujutsu,
respectively).
|
2304.10036 | Benjamin Ramtoula | Benjamin Ramtoula, Matthew Gadd, Paul Newman, Daniele De Martini | Visual DNA: Representing and Comparing Images using Distributions of
Neuron Activations | Published at CVPR 2023. Project page with code:
https://bramtoula.github.io/vdna/ | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Selecting appropriate datasets is critical in modern computer vision.
However, no general-purpose tools exist to evaluate the extent to which two
datasets differ. For this, we propose representing images - and by extension
datasets - using Distributions of Neuron Activations (DNAs). DNAs fit
distributions, such as histograms or Gaussians, to activations of neurons in a
pre-trained feature extractor through which we pass the image(s) to represent.
This extractor is frozen for all datasets, and we rely on its generally
expressive power in feature space. By comparing two DNAs, we can evaluate the
extent to which two datasets differ with granular control over the comparison
attributes of interest, providing the ability to customise the way distances
are measured to suit the requirements of the task at hand. Furthermore, DNAs
are compact, representing datasets of any size with less than 15 megabytes. We
demonstrate the value of DNAs by evaluating their applicability on several
tasks, including conditional dataset comparison, synthetic image evaluation,
and transfer learning, and across diverse datasets, ranging from synthetic cat
images to celebrity faces and urban driving scenes.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 01:23:36 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Ramtoula",
"Benjamin",
""
],
[
"Gadd",
"Matthew",
""
],
[
"Newman",
"Paul",
""
],
[
"De Martini",
"Daniele",
""
]
] | TITLE: Visual DNA: Representing and Comparing Images using Distributions of
Neuron Activations
ABSTRACT: Selecting appropriate datasets is critical in modern computer vision.
However, no general-purpose tools exist to evaluate the extent to which two
datasets differ. For this, we propose representing images - and by extension
datasets - using Distributions of Neuron Activations (DNAs). DNAs fit
distributions, such as histograms or Gaussians, to activations of neurons in a
pre-trained feature extractor through which we pass the image(s) to represent.
This extractor is frozen for all datasets, and we rely on its generally
expressive power in feature space. By comparing two DNAs, we can evaluate the
extent to which two datasets differ with granular control over the comparison
attributes of interest, providing the ability to customise the way distances
are measured to suit the requirements of the task at hand. Furthermore, DNAs
are compact, representing datasets of any size with less than 15 megabytes. We
demonstrate the value of DNAs by evaluating their applicability on several
tasks, including conditional dataset comparison, synthetic image evaluation,
and transfer learning, and across diverse datasets, ranging from synthetic cat
images to celebrity faces and urban driving scenes.
|
2304.10044 | Zahra Mirzamomen Mrs | Zahra Mirzamomen, Marcel B\"ohme | Finding Bug-Inducing Program Environments | null | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | Some bugs cannot be exposed by program inputs, but only by certain program
environments. During execution, most programs access various resources, like
databases, files, or devices, that are external to the program and thus part of
the program's environment. In this paper, we present a coverage-guided,
mutation-based environment synthesis approach of bug-inducing program
environments. Specifically, we observe that programs interact with their
environment via dedicated system calls and propose to intercept these system
calls (i) to capture the resources accessed during the first execution of an
input as initial program environment, and (ii) mutate copies of these resources
during subsequent executions of that input to generate slightly changed program
environments. Any generated environment that is observed to increase coverage
is added to the corpus of environment seeds and becomes subject to further
fuzzing. Bug-inducing program environments are reported to the user.
Experiments demonstrate the effectiveness of our approach. We implemented a
prototype called AFLChaos which found bugs in the resource-handling code of
five (5) of the seven (7) open source projects in our benchmark set (incl.
OpenSSL). Automatically, AFLChaos generated environments consisting of
bug-inducing databases used for storing information, bug-inducing multimedia
files used for streaming, bug-inducing cryptographic keys used for encryption,
and bug-inducing configuration files used to configure the program. To support
open science, we publish the experimental infrastructure, our tool, and all
data.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 01:46:19 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Mirzamomen",
"Zahra",
""
],
[
"Böhme",
"Marcel",
""
]
] | TITLE: Finding Bug-Inducing Program Environments
ABSTRACT: Some bugs cannot be exposed by program inputs, but only by certain program
environments. During execution, most programs access various resources, like
databases, files, or devices, that are external to the program and thus part of
the program's environment. In this paper, we present a coverage-guided,
mutation-based environment synthesis approach of bug-inducing program
environments. Specifically, we observe that programs interact with their
environment via dedicated system calls and propose to intercept these system
calls (i) to capture the resources accessed during the first execution of an
input as initial program environment, and (ii) mutate copies of these resources
during subsequent executions of that input to generate slightly changed program
environments. Any generated environment that is observed to increase coverage
is added to the corpus of environment seeds and becomes subject to further
fuzzing. Bug-inducing program environments are reported to the user.
Experiments demonstrate the effectiveness of our approach. We implemented a
prototype called AFLChaos which found bugs in the resource-handling code of
five (5) of the seven (7) open source projects in our benchmark set (incl.
OpenSSL). Automatically, AFLChaos generated environments consisting of
bug-inducing databases used for storing information, bug-inducing multimedia
files used for streaming, bug-inducing cryptographic keys used for encryption,
and bug-inducing configuration files used to configure the program. To support
open science, we publish the experimental infrastructure, our tool, and all
data.
|
2304.10051 | Ss Z | Hui Dou, Shanshan Zhu, Yiwen Zhang, Pengfei Chen and Zibin Zheng | HyperTuner: A Cross-Layer Multi-Objective Hyperparameter Auto-Tuning
Framework for Data Analytic Services | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hyper-parameters optimization (HPO) is vital for machine learning models.
Besides model accuracy, other tuning intentions such as model training time and
energy consumption are also worthy of attention from data analytic service
providers. Hence, it is essential to take both model hyperparameters and system
parameters into consideration to execute cross-layer multi-objective
hyperparameter auto-tuning. Towards this challenging target, we propose
HyperTuner in this paper. To address the formulated high-dimensional black-box
multi-objective optimization problem, HyperTuner first conducts multi-objective
parameter importance ranking with its MOPIR algorithm and then leverages the
proposed ADUMBO algorithm to find the Pareto-optimal configuration set. During
each iteration, ADUMBO selects the most promising configuration from the
generated Pareto candidate set via maximizing a new well-designed metric, which
can adaptively leverage the uncertainty as well as the predicted mean across
all the surrogate models along with the iteration times. We evaluate HyperTuner
on our local distributed TensorFlow cluster and experimental results show that
it is always able to find a better Pareto configuration front superior in both
convergence and diversity compared with the other four baseline algorithms.
Besides, experiments with different training datasets, different optimization
objectives and different machine learning platforms verify that HyperTuner can
well adapt to various data analytic service scenarios.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 02:19:10 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Dou",
"Hui",
""
],
[
"Zhu",
"Shanshan",
""
],
[
"Zhang",
"Yiwen",
""
],
[
"Chen",
"Pengfei",
""
],
[
"Zheng",
"Zibin",
""
]
] | TITLE: HyperTuner: A Cross-Layer Multi-Objective Hyperparameter Auto-Tuning
Framework for Data Analytic Services
ABSTRACT: Hyper-parameters optimization (HPO) is vital for machine learning models.
Besides model accuracy, other tuning intentions such as model training time and
energy consumption are also worthy of attention from data analytic service
providers. Hence, it is essential to take both model hyperparameters and system
parameters into consideration to execute cross-layer multi-objective
hyperparameter auto-tuning. Towards this challenging target, we propose
HyperTuner in this paper. To address the formulated high-dimensional black-box
multi-objective optimization problem, HyperTuner first conducts multi-objective
parameter importance ranking with its MOPIR algorithm and then leverages the
proposed ADUMBO algorithm to find the Pareto-optimal configuration set. During
each iteration, ADUMBO selects the most promising configuration from the
generated Pareto candidate set via maximizing a new well-designed metric, which
can adaptively leverage the uncertainty as well as the predicted mean across
all the surrogate models along with the iteration times. We evaluate HyperTuner
on our local distributed TensorFlow cluster and experimental results show that
it is always able to find a better Pareto configuration front superior in both
convergence and diversity compared with the other four baseline algorithms.
Besides, experiments with different training datasets, different optimization
objectives and different machine learning platforms verify that HyperTuner can
well adapt to various data analytic service scenarios.
|
2304.10054 | Zhuoran Zheng | Zhuoran Zheng and Xiuyi Jia | Complex Mixer for MedMNIST Classification Decathlon | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | With the development of the medical image field, researchers seek to develop
a class of datasets to block the need for medical knowledge, such as
\text{MedMNIST} (v2). MedMNIST (v2) includes a large number of small-sized (28
$\times$ 28 or 28 $\times$ 28 $\times$ 28) medical samples and the
corresponding expert annotations (class label). The existing baseline model
(Google AutoML Vision, ResNet-50+3D) can reach an average accuracy of over 70\%
on MedMNIST (v2) datasets, which is comparable to the performance of expert
decision-making. Nevertheless, we note that there are two insurmountable
obstacles to modeling on MedMNIST (v2): 1) the raw images are cropped to low
scales may cause effective recognition information to be dropped and the
classifier to have difficulty in tracing accurate decision boundaries; 2) the
labelers' subjective insight may cause many uncertainties in the label space.
To address these issues, we develop a Complex Mixer (C-Mixer) with a
pre-training framework to alleviate the problem of insufficient information and
uncertainty in the label space by introducing an incentive imaginary matrix and
a self-supervised scheme with random masking. Our method (incentive learning
and self-supervised learning with masking) shows surprising potential on both
the standard MedMNIST (v2) dataset, the customized weakly supervised datasets,
and other image enhancement tasks.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 02:34:36 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Zheng",
"Zhuoran",
""
],
[
"Jia",
"Xiuyi",
""
]
] | TITLE: Complex Mixer for MedMNIST Classification Decathlon
ABSTRACT: With the development of the medical image field, researchers seek to develop
a class of datasets to block the need for medical knowledge, such as
\text{MedMNIST} (v2). MedMNIST (v2) includes a large number of small-sized (28
$\times$ 28 or 28 $\times$ 28 $\times$ 28) medical samples and the
corresponding expert annotations (class label). The existing baseline model
(Google AutoML Vision, ResNet-50+3D) can reach an average accuracy of over 70\%
on MedMNIST (v2) datasets, which is comparable to the performance of expert
decision-making. Nevertheless, we note that there are two insurmountable
obstacles to modeling on MedMNIST (v2): 1) the raw images are cropped to low
scales may cause effective recognition information to be dropped and the
classifier to have difficulty in tracing accurate decision boundaries; 2) the
labelers' subjective insight may cause many uncertainties in the label space.
To address these issues, we develop a Complex Mixer (C-Mixer) with a
pre-training framework to alleviate the problem of insufficient information and
uncertainty in the label space by introducing an incentive imaginary matrix and
a self-supervised scheme with random masking. Our method (incentive learning
and self-supervised learning with masking) shows surprising potential on both
the standard MedMNIST (v2) dataset, the customized weakly supervised datasets,
and other image enhancement tasks.
|
2304.10066 | Jacky Chen Long Chai | Jacky Chen Long Chai, Tiong-Sik Ng, Cheng-Yaw Low, Jaewoo Park, Andrew
Beng Jin Teoh | Recognizability Embedding Enhancement for Very Low-Resolution Face
Recognition and Quality Estimation | Accepted to CVPR23 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Very low-resolution face recognition (VLRFR) poses unique challenges, such as
tiny regions of interest and poor resolution due to extreme standoff distance
or wide viewing angle of the acquisition devices. In this paper, we study
principled approaches to elevate the recognizability of a face in the embedding
space instead of the visual quality. We first formulate a robust learning-based
face recognizability measure, namely recognizability index (RI), based on two
criteria: (i) proximity of each face embedding against the unrecognizable faces
cluster center and (ii) closeness of each face embedding against its positive
and negative class prototypes. We then devise an index diversion loss to push
the hard-to-recognize face embedding with low RI away from unrecognizable faces
cluster to boost the RI, which reflects better recognizability. Additionally, a
perceptibility attention mechanism is introduced to attend to the most
recognizable face regions, which offers better explanatory and discriminative
traits for embedding learning. Our proposed model is trained end-to-end and
simultaneously serves recognizability-aware embedding learning and face quality
estimation. To address VLRFR, our extensive evaluations on three challenging
low-resolution datasets and face quality assessment demonstrate the superiority
of the proposed model over the state-of-the-art methods.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 03:18:03 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Chai",
"Jacky Chen Long",
""
],
[
"Ng",
"Tiong-Sik",
""
],
[
"Low",
"Cheng-Yaw",
""
],
[
"Park",
"Jaewoo",
""
],
[
"Teoh",
"Andrew Beng Jin",
""
]
] | TITLE: Recognizability Embedding Enhancement for Very Low-Resolution Face
Recognition and Quality Estimation
ABSTRACT: Very low-resolution face recognition (VLRFR) poses unique challenges, such as
tiny regions of interest and poor resolution due to extreme standoff distance
or wide viewing angle of the acquisition devices. In this paper, we study
principled approaches to elevate the recognizability of a face in the embedding
space instead of the visual quality. We first formulate a robust learning-based
face recognizability measure, namely recognizability index (RI), based on two
criteria: (i) proximity of each face embedding against the unrecognizable faces
cluster center and (ii) closeness of each face embedding against its positive
and negative class prototypes. We then devise an index diversion loss to push
the hard-to-recognize face embedding with low RI away from unrecognizable faces
cluster to boost the RI, which reflects better recognizability. Additionally, a
perceptibility attention mechanism is introduced to attend to the most
recognizable face regions, which offers better explanatory and discriminative
traits for embedding learning. Our proposed model is trained end-to-end and
simultaneously serves recognizability-aware embedding learning and face quality
estimation. To address VLRFR, our extensive evaluations on three challenging
low-resolution datasets and face quality assessment demonstrate the superiority
of the proposed model over the state-of-the-art methods.
|
2304.10068 | Joel Dabrowski Dr | Joel Janek Dabrowski and Ashfaqur Rahman | Fruit Picker Activity Recognition with Wearable Sensors and Machine
Learning | Accepted at IEEE International Joint Conference on Neural Networks
(IJCNN) conference, 2023 | null | null | null | cs.LG eess.SP | http://creativecommons.org/licenses/by/4.0/ | In this paper we present a novel application of detecting fruit picker
activities based on time series data generated from wearable sensors. During
harvesting, fruit pickers pick fruit into wearable bags and empty these bags
into harvesting bins located in the orchard. Once full, these bins are quickly
transported to a cooled pack house to improve the shelf life of picked fruits.
For farmers and managers, the knowledge of when a picker bag is emptied is
important for managing harvesting bins more effectively to minimise the time
the picked fruit is left out in the heat (resulting in reduced shelf life). We
propose a means to detect these bag-emptying events using human activity
recognition with wearable sensors and machine learning methods. We develop a
semi-supervised approach to labelling the data. A feature-based machine
learning ensemble model and a deep recurrent convolutional neural network are
developed and tested on a real-world dataset. When compared, the neural network
achieves 86% detection accuracy.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 03:38:08 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Dabrowski",
"Joel Janek",
""
],
[
"Rahman",
"Ashfaqur",
""
]
] | TITLE: Fruit Picker Activity Recognition with Wearable Sensors and Machine
Learning
ABSTRACT: In this paper we present a novel application of detecting fruit picker
activities based on time series data generated from wearable sensors. During
harvesting, fruit pickers pick fruit into wearable bags and empty these bags
into harvesting bins located in the orchard. Once full, these bins are quickly
transported to a cooled pack house to improve the shelf life of picked fruits.
For farmers and managers, the knowledge of when a picker bag is emptied is
important for managing harvesting bins more effectively to minimise the time
the picked fruit is left out in the heat (resulting in reduced shelf life). We
propose a means to detect these bag-emptying events using human activity
recognition with wearable sensors and machine learning methods. We develop a
semi-supervised approach to labelling the data. A feature-based machine
learning ensemble model and a deep recurrent convolutional neural network are
developed and tested on a real-world dataset. When compared, the neural network
achieves 86% detection accuracy.
|
2304.10080 | Lin Gao | Yu-Tao Liu, Li Wang, Jie yang, Weikai Chen, Xiaoxu Meng, Bo Yang, Lin
Gao | NeUDF: Leaning Neural Unsigned Distance Fields with Volume Rendering | Accepted by CVPR 2023, 24 pages, project:
http://geometrylearning.com/neudf/ | null | null | null | cs.CV cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-view shape reconstruction has achieved impressive progresses thanks to
the latest advances in neural implicit surface rendering. However, existing
methods based on signed distance function (SDF) are limited to closed surfaces,
failing to reconstruct a wide range of real-world objects that contain
open-surface structures. In this work, we introduce a new neural rendering
framework, coded NeUDF, that can reconstruct surfaces with arbitrary topologies
solely from multi-view supervision. To gain the flexibility of representing
arbitrary surfaces, NeUDF leverages the unsigned distance function (UDF) as
surface representation. While a naive extension of an SDF-based neural renderer
cannot scale to UDF, we propose two new formulations of weight function
specially tailored for UDF-based volume rendering. Furthermore, to cope with
open surface rendering, where the in/out test is no longer valid, we present a
dedicated normal regularization strategy to resolve the surface orientation
ambiguity. We extensively evaluate our method over a number of challenging
datasets, including DTU}, MGN, and Deep Fashion 3D. Experimental results
demonstrate that nEudf can significantly outperform the state-of-the-art method
in the task of multi-view surface reconstruction, especially for complex shapes
with open boundaries.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 04:14:42 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Liu",
"Yu-Tao",
""
],
[
"Wang",
"Li",
""
],
[
"yang",
"Jie",
""
],
[
"Chen",
"Weikai",
""
],
[
"Meng",
"Xiaoxu",
""
],
[
"Yang",
"Bo",
""
],
[
"Gao",
"Lin",
""
]
] | TITLE: NeUDF: Leaning Neural Unsigned Distance Fields with Volume Rendering
ABSTRACT: Multi-view shape reconstruction has achieved impressive progresses thanks to
the latest advances in neural implicit surface rendering. However, existing
methods based on signed distance function (SDF) are limited to closed surfaces,
failing to reconstruct a wide range of real-world objects that contain
open-surface structures. In this work, we introduce a new neural rendering
framework, coded NeUDF, that can reconstruct surfaces with arbitrary topologies
solely from multi-view supervision. To gain the flexibility of representing
arbitrary surfaces, NeUDF leverages the unsigned distance function (UDF) as
surface representation. While a naive extension of an SDF-based neural renderer
cannot scale to UDF, we propose two new formulations of weight function
specially tailored for UDF-based volume rendering. Furthermore, to cope with
open surface rendering, where the in/out test is no longer valid, we present a
dedicated normal regularization strategy to resolve the surface orientation
ambiguity. We extensively evaluate our method over a number of challenging
datasets, including DTU}, MGN, and Deep Fashion 3D. Experimental results
demonstrate that nEudf can significantly outperform the state-of-the-art method
in the task of multi-view surface reconstruction, especially for complex shapes
with open boundaries.
|
2304.10091 | Xiao Wang | Jun Zhu, Jiandong Jin, Zihan Yang, Xiaohao Wu, Xiao Wang | Learning CLIP Guided Visual-Text Fusion Transformer for Video-based
Pedestrian Attribute Recognition | Accepted by CVPR-2023 Workshop@NFVLR (New Frontiers in Visual
Language Reasoning: Compositionality, Prompts and Causality) | null | null | null | cs.CV cs.MM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing pedestrian attribute recognition (PAR) algorithms are mainly
developed based on a static image. However, the performance is not reliable for
images with challenging factors, such as heavy occlusion, motion blur, etc. In
this work, we propose to understand human attributes using video frames that
can make full use of temporal information. Specifically, we formulate the
video-based PAR as a vision-language fusion problem and adopt pre-trained big
models CLIP to extract the feature embeddings of given video frames. To better
utilize the semantic information, we take the attribute list as another input
and transform the attribute words/phrase into the corresponding sentence via
split, expand, and prompt. Then, the text encoder of CLIP is utilized for
language embedding. The averaged visual tokens and text tokens are concatenated
and fed into a fusion Transformer for multi-modal interactive learning. The
enhanced tokens will be fed into a classification head for pedestrian attribute
prediction. Extensive experiments on a large-scale video-based PAR dataset
fully validated the effectiveness of our proposed framework.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 05:18:28 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Zhu",
"Jun",
""
],
[
"Jin",
"Jiandong",
""
],
[
"Yang",
"Zihan",
""
],
[
"Wu",
"Xiaohao",
""
],
[
"Wang",
"Xiao",
""
]
] | TITLE: Learning CLIP Guided Visual-Text Fusion Transformer for Video-based
Pedestrian Attribute Recognition
ABSTRACT: Existing pedestrian attribute recognition (PAR) algorithms are mainly
developed based on a static image. However, the performance is not reliable for
images with challenging factors, such as heavy occlusion, motion blur, etc. In
this work, we propose to understand human attributes using video frames that
can make full use of temporal information. Specifically, we formulate the
video-based PAR as a vision-language fusion problem and adopt pre-trained big
models CLIP to extract the feature embeddings of given video frames. To better
utilize the semantic information, we take the attribute list as another input
and transform the attribute words/phrase into the corresponding sentence via
split, expand, and prompt. Then, the text encoder of CLIP is utilized for
language embedding. The averaged visual tokens and text tokens are concatenated
and fed into a fusion Transformer for multi-modal interactive learning. The
enhanced tokens will be fed into a classification head for pedestrian attribute
prediction. Extensive experiments on a large-scale video-based PAR dataset
fully validated the effectiveness of our proposed framework.
|
2304.10097 | Fuxiang Yang | Tonghua Su, Fuxiang Yang, Xiang Zhou, Donglin Di, Zhongjie Wang,
Songze Li | Scene Style Text Editing | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we propose a task called "Scene Style Text Editing (SSTE)",
changing the text content as well as the text style of the source image while
keeping the original text scene. Existing methods neglect to fine-grained
adjust the style of the foreground text, such as its rotation angle, color, and
font type. To tackle this task, we propose a quadruple framework named
"QuadNet" to embed and adjust foreground text styles in the latent feature
space. Specifically, QuadNet consists of four parts, namely background
inpainting, style encoder, content encoder, and fusion generator. The
background inpainting erases the source text content and recovers the
appropriate background with a highly authentic texture. The style encoder
extracts the style embedding of the foreground text. The content encoder
provides target text representations in the latent feature space to implement
the content edits. The fusion generator combines the information yielded from
the mentioned parts and generates the rendered text images. Practically, our
method is capable of performing promisingly on real-world datasets with merely
string-level annotation. To the best of our knowledge, our work is the first to
finely manipulate the foreground text content and style by deeply semantic
editing in the latent feature space. Extensive experiments demonstrate that
QuadNet has the ability to generate photo-realistic foreground text and avoid
source text shadows in real-world scenes when editing text content.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 05:36:49 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Su",
"Tonghua",
""
],
[
"Yang",
"Fuxiang",
""
],
[
"Zhou",
"Xiang",
""
],
[
"Di",
"Donglin",
""
],
[
"Wang",
"Zhongjie",
""
],
[
"Li",
"Songze",
""
]
] | TITLE: Scene Style Text Editing
ABSTRACT: In this work, we propose a task called "Scene Style Text Editing (SSTE)",
changing the text content as well as the text style of the source image while
keeping the original text scene. Existing methods neglect to fine-grained
adjust the style of the foreground text, such as its rotation angle, color, and
font type. To tackle this task, we propose a quadruple framework named
"QuadNet" to embed and adjust foreground text styles in the latent feature
space. Specifically, QuadNet consists of four parts, namely background
inpainting, style encoder, content encoder, and fusion generator. The
background inpainting erases the source text content and recovers the
appropriate background with a highly authentic texture. The style encoder
extracts the style embedding of the foreground text. The content encoder
provides target text representations in the latent feature space to implement
the content edits. The fusion generator combines the information yielded from
the mentioned parts and generates the rendered text images. Practically, our
method is capable of performing promisingly on real-world datasets with merely
string-level annotation. To the best of our knowledge, our work is the first to
finely manipulate the foreground text content and style by deeply semantic
editing in the latent feature space. Extensive experiments demonstrate that
QuadNet has the ability to generate photo-realistic foreground text and avoid
source text shadows in real-world scenes when editing text content.
|
2304.10113 | Goirik Chakrabarty | Goirik Chakrabarty, Manogna Sreenivas and Soma Biswas | SATA: Source Anchoring and Target Alignment Network for Continual Test
Time Adaptation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adapting a trained model to perform satisfactorily on continually changing
testing domains/environments is an important and challenging task. In this
work, we propose a novel framework, SATA, which aims to satisfy the following
characteristics required for online adaptation: 1) can work seamlessly with
different (preferably small) batch sizes to reduce latency; 2) should continue
to work well for the source domain; 3) should have minimal tunable
hyper-parameters and storage requirements. Given a pre-trained network trained
on source domain data, the proposed SATA framework modifies the batch-norm
affine parameters using source anchoring based self-distillation. This ensures
that the model incorporates the knowledge of the newly encountered domains,
without catastrophically forgetting about the previously seen ones. We also
propose a source-prototype driven contrastive alignment to ensure natural
grouping of the target samples, while maintaining the already learnt semantic
information. Extensive evaluation on three benchmark datasets under challenging
settings justify the effectiveness of SATA for real-world applications.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 06:38:33 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Chakrabarty",
"Goirik",
""
],
[
"Sreenivas",
"Manogna",
""
],
[
"Biswas",
"Soma",
""
]
] | TITLE: SATA: Source Anchoring and Target Alignment Network for Continual Test
Time Adaptation
ABSTRACT: Adapting a trained model to perform satisfactorily on continually changing
testing domains/environments is an important and challenging task. In this
work, we propose a novel framework, SATA, which aims to satisfy the following
characteristics required for online adaptation: 1) can work seamlessly with
different (preferably small) batch sizes to reduce latency; 2) should continue
to work well for the source domain; 3) should have minimal tunable
hyper-parameters and storage requirements. Given a pre-trained network trained
on source domain data, the proposed SATA framework modifies the batch-norm
affine parameters using source anchoring based self-distillation. This ensures
that the model incorporates the knowledge of the newly encountered domains,
without catastrophically forgetting about the previously seen ones. We also
propose a source-prototype driven contrastive alignment to ensure natural
grouping of the target samples, while maintaining the already learnt semantic
information. Extensive evaluation on three benchmark datasets under challenging
settings justify the effectiveness of SATA for real-world applications.
|
2304.10137 | Ahsan Tanveer | Ahsan Tanveer, Sarvat Mushtaq Ahmad | Genetic-Algorithm-Based Proportional Integral Controller (GAPI) for ROV
Steering Control | 6 pages, 3 figures, presented at the 2nd International Conference on
Emerging Trends in Electronic and Telecommunication Engineering (INTERACT),
published in Engineering Proceedings 32 | Engineering Proceedings 32, no. 1: 4, 2023 | 10.3390/engproc2023032004 | null | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | This article presents the design and real-time implementation of an optimal
controller for precise steering control of a remotely operated underwater
vehicle (ROV). A PI controller is investigated to achieve the desired steering
performance. The gain parameters of the controller are tuned using the genetic
algorithm (GA). The experimental response corresponding to the step waveform
for the GA is obtained. A root-locus-tuned PI controller alongside a
simulated-annealing-based PI controller (SAPI) is used to benchmark the
response characteristics such as overshoot, peak time, and settling time. The
experimental findings indicate that GAPI provides considerably better
performance than SAPI and the root-locus-tuned controller.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 07:51:46 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Tanveer",
"Ahsan",
""
],
[
"Ahmad",
"Sarvat Mushtaq",
""
]
] | TITLE: Genetic-Algorithm-Based Proportional Integral Controller (GAPI) for ROV
Steering Control
ABSTRACT: This article presents the design and real-time implementation of an optimal
controller for precise steering control of a remotely operated underwater
vehicle (ROV). A PI controller is investigated to achieve the desired steering
performance. The gain parameters of the controller are tuned using the genetic
algorithm (GA). The experimental response corresponding to the step waveform
for the GA is obtained. A root-locus-tuned PI controller alongside a
simulated-annealing-based PI controller (SAPI) is used to benchmark the
response characteristics such as overshoot, peak time, and settling time. The
experimental findings indicate that GAPI provides considerably better
performance than SAPI and the root-locus-tuned controller.
|
2304.10177 | Zhicheng Sun | Zhicheng Sun, Yadong Mu, Gang Hua | Regularizing Second-Order Influences for Continual Learning | CVPR 2023 | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Continual learning aims to learn on non-stationary data streams without
catastrophically forgetting previous knowledge. Prevalent replay-based methods
address this challenge by rehearsing on a small buffer holding the seen data,
for which a delicate sample selection strategy is required. However, existing
selection schemes typically seek only to maximize the utility of the ongoing
selection, overlooking the interference between successive rounds of selection.
Motivated by this, we dissect the interaction of sequential selection steps
within a framework built on influence functions. We manage to identify a new
class of second-order influences that will gradually amplify incidental bias in
the replay buffer and compromise the selection process. To regularize the
second-order effects, a novel selection objective is proposed, which also has
clear connections to two widely adopted criteria. Furthermore, we present an
efficient implementation for optimizing the proposed criterion. Experiments on
multiple continual learning benchmarks demonstrate the advantage of our
approach over state-of-the-art methods. Code is available at
https://github.com/feifeiobama/InfluenceCL.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 09:30:35 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Sun",
"Zhicheng",
""
],
[
"Mu",
"Yadong",
""
],
[
"Hua",
"Gang",
""
]
] | TITLE: Regularizing Second-Order Influences for Continual Learning
ABSTRACT: Continual learning aims to learn on non-stationary data streams without
catastrophically forgetting previous knowledge. Prevalent replay-based methods
address this challenge by rehearsing on a small buffer holding the seen data,
for which a delicate sample selection strategy is required. However, existing
selection schemes typically seek only to maximize the utility of the ongoing
selection, overlooking the interference between successive rounds of selection.
Motivated by this, we dissect the interaction of sequential selection steps
within a framework built on influence functions. We manage to identify a new
class of second-order influences that will gradually amplify incidental bias in
the replay buffer and compromise the selection process. To regularize the
second-order effects, a novel selection objective is proposed, which also has
clear connections to two widely adopted criteria. Furthermore, we present an
efficient implementation for optimizing the proposed criterion. Experiments on
multiple continual learning benchmarks demonstrate the advantage of our
approach over state-of-the-art methods. Code is available at
https://github.com/feifeiobama/InfluenceCL.
|
2304.10195 | Guangyuan Ma | Guangyuan Ma, Xing Wu, Peng Wang, Songlin Hu | CoT-MoTE: Exploring ConTextual Masked Auto-Encoder Pre-training with
Mixture-of-Textual-Experts for Passage Retrieval | null | null | null | null | cs.CL cs.IR | http://creativecommons.org/licenses/by/4.0/ | Passage retrieval aims to retrieve relevant passages from large collections
of the open-domain corpus. Contextual Masked Auto-Encoding has been proven
effective in representation bottleneck pre-training of a monolithic
dual-encoder for passage retrieval. Siamese or fully separated dual-encoders
are often adopted as basic retrieval architecture in the pre-training and
fine-tuning stages for encoding queries and passages into their latent
embedding spaces. However, simply sharing or separating the parameters of the
dual-encoder results in an imbalanced discrimination of the embedding spaces.
In this work, we propose to pre-train Contextual Masked Auto-Encoder with
Mixture-of-Textual-Experts (CoT-MoTE). Specifically, we incorporate
textual-specific experts for individually encoding the distinct properties of
queries and passages. Meanwhile, a shared self-attention layer is still kept
for unified attention modeling. Results on large-scale passage retrieval
benchmarks show steady improvement in retrieval performances. The quantitive
analysis also shows a more balanced discrimination of the latent embedding
spaces.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 10:12:09 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Ma",
"Guangyuan",
""
],
[
"Wu",
"Xing",
""
],
[
"Wang",
"Peng",
""
],
[
"Hu",
"Songlin",
""
]
] | TITLE: CoT-MoTE: Exploring ConTextual Masked Auto-Encoder Pre-training with
Mixture-of-Textual-Experts for Passage Retrieval
ABSTRACT: Passage retrieval aims to retrieve relevant passages from large collections
of the open-domain corpus. Contextual Masked Auto-Encoding has been proven
effective in representation bottleneck pre-training of a monolithic
dual-encoder for passage retrieval. Siamese or fully separated dual-encoders
are often adopted as basic retrieval architecture in the pre-training and
fine-tuning stages for encoding queries and passages into their latent
embedding spaces. However, simply sharing or separating the parameters of the
dual-encoder results in an imbalanced discrimination of the embedding spaces.
In this work, we propose to pre-train Contextual Masked Auto-Encoder with
Mixture-of-Textual-Experts (CoT-MoTE). Specifically, we incorporate
textual-specific experts for individually encoding the distinct properties of
queries and passages. Meanwhile, a shared self-attention layer is still kept
for unified attention modeling. Results on large-scale passage retrieval
benchmarks show steady improvement in retrieval performances. The quantitive
analysis also shows a more balanced discrimination of the latent embedding
spaces.
|
2304.10199 | Yuyuan Li | Yuyuan Li, Chaochao Chen, Xiaolin Zheng, Yizhao Zhang, Biao Gong, Jun
Wang | Selective and Collaborative Influence Function for Efficient
Recommendation Unlearning | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent regulations on the Right to be Forgotten have greatly influenced the
way of running a recommender system, because users now have the right to
withdraw their private data. Besides simply deleting the target data in the
database, unlearning the associated data lineage e.g., the learned personal
features and preferences in the model, is also necessary for data withdrawal.
Existing unlearning methods are mainly devised for generalized machine learning
models in classification tasks. In this paper, we first identify two main
disadvantages of directly applying existing unlearning methods in the context
of recommendation, i.e., (i) unsatisfactory efficiency for large-scale
recommendation models and (ii) destruction of collaboration across users and
items. To tackle the above issues, we propose an extra-efficient recommendation
unlearning method based on Selective and Collaborative Influence Function
(SCIF). Our proposed method can (i) avoid any kind of retraining which is
computationally prohibitive for large-scale systems, (ii) further enhance
efficiency by selectively updating user embedding and (iii) preserve the
collaboration across the remaining users and items. Furthermore, in order to
evaluate the unlearning completeness, we define a Membership Inference Oracle
(MIO), which can justify whether the unlearned data points were in the training
set of the model, i.e., whether a data point was completely unlearned.
Extensive experiments on two benchmark datasets demonstrate that our proposed
method can not only greatly enhance unlearning efficiency, but also achieve
adequate unlearning completeness. More importantly, our proposed method
outperforms the state-of-the-art unlearning method regarding comprehensive
recommendation metrics.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 10:35:34 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Li",
"Yuyuan",
""
],
[
"Chen",
"Chaochao",
""
],
[
"Zheng",
"Xiaolin",
""
],
[
"Zhang",
"Yizhao",
""
],
[
"Gong",
"Biao",
""
],
[
"Wang",
"Jun",
""
]
] | TITLE: Selective and Collaborative Influence Function for Efficient
Recommendation Unlearning
ABSTRACT: Recent regulations on the Right to be Forgotten have greatly influenced the
way of running a recommender system, because users now have the right to
withdraw their private data. Besides simply deleting the target data in the
database, unlearning the associated data lineage e.g., the learned personal
features and preferences in the model, is also necessary for data withdrawal.
Existing unlearning methods are mainly devised for generalized machine learning
models in classification tasks. In this paper, we first identify two main
disadvantages of directly applying existing unlearning methods in the context
of recommendation, i.e., (i) unsatisfactory efficiency for large-scale
recommendation models and (ii) destruction of collaboration across users and
items. To tackle the above issues, we propose an extra-efficient recommendation
unlearning method based on Selective and Collaborative Influence Function
(SCIF). Our proposed method can (i) avoid any kind of retraining which is
computationally prohibitive for large-scale systems, (ii) further enhance
efficiency by selectively updating user embedding and (iii) preserve the
collaboration across the remaining users and items. Furthermore, in order to
evaluate the unlearning completeness, we define a Membership Inference Oracle
(MIO), which can justify whether the unlearned data points were in the training
set of the model, i.e., whether a data point was completely unlearned.
Extensive experiments on two benchmark datasets demonstrate that our proposed
method can not only greatly enhance unlearning efficiency, but also achieve
adequate unlearning completeness. More importantly, our proposed method
outperforms the state-of-the-art unlearning method regarding comprehensive
recommendation metrics.
|
2304.10211 | Sami Barchid | Sami Barchid, Benjamin Allaert, Amel Aissaoui, Jos\'e Mennesson,
Chaabane Dj\'eraba | Spiking-Fer: Spiking Neural Network for Facial Expression Recognition
With Event Cameras | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Facial Expression Recognition (FER) is an active research domain that has
shown great progress recently, notably thanks to the use of large deep learning
models. However, such approaches are particularly energy intensive, which makes
their deployment difficult for edge devices. To address this issue, Spiking
Neural Networks (SNNs) coupled with event cameras are a promising alternative,
capable of processing sparse and asynchronous events with lower energy
consumption. In this paper, we establish the first use of event cameras for
FER, named "Event-based FER", and propose the first related benchmarks by
converting popular video FER datasets to event streams. To deal with this new
task, we propose "Spiking-FER", a deep convolutional SNN model, and compare it
against a similar Artificial Neural Network (ANN). Experiments show that the
proposed approach achieves comparable performance to the ANN architecture,
while consuming less energy by orders of magnitude (up to 65.39x). In addition,
an experimental study of various event-based data augmentation techniques is
performed to provide insights into the efficient transformations specific to
event-based FER.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 10:59:56 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Barchid",
"Sami",
""
],
[
"Allaert",
"Benjamin",
""
],
[
"Aissaoui",
"Amel",
""
],
[
"Mennesson",
"José",
""
],
[
"Djéraba",
"Chaabane",
""
]
] | TITLE: Spiking-Fer: Spiking Neural Network for Facial Expression Recognition
With Event Cameras
ABSTRACT: Facial Expression Recognition (FER) is an active research domain that has
shown great progress recently, notably thanks to the use of large deep learning
models. However, such approaches are particularly energy intensive, which makes
their deployment difficult for edge devices. To address this issue, Spiking
Neural Networks (SNNs) coupled with event cameras are a promising alternative,
capable of processing sparse and asynchronous events with lower energy
consumption. In this paper, we establish the first use of event cameras for
FER, named "Event-based FER", and propose the first related benchmarks by
converting popular video FER datasets to event streams. To deal with this new
task, we propose "Spiking-FER", a deep convolutional SNN model, and compare it
against a similar Artificial Neural Network (ANN). Experiments show that the
proposed approach achieves comparable performance to the ANN architecture,
while consuming less energy by orders of magnitude (up to 65.39x). In addition,
an experimental study of various event-based data augmentation techniques is
performed to provide insights into the efficient transformations specific to
event-based FER.
|
2304.10216 | Yusser Al Ghussin | Yusser Al Ghussin, Jingyi Zhang, Josef van Genabith | Exploring Paracrawl for Document-level Neural Machine Translation | Accepted to EACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Document-level neural machine translation (NMT) has outperformed
sentence-level NMT on a number of datasets. However, document-level NMT is
still not widely adopted in real-world translation systems mainly due to the
lack of large-scale general-domain training data for document-level NMT. We
examine the effectiveness of using Paracrawl for learning document-level
translation. Paracrawl is a large-scale parallel corpus crawled from the
Internet and contains data from various domains. The official Paracrawl corpus
was released as parallel sentences (extracted from parallel webpages) and
therefore previous works only used Paracrawl for learning sentence-level
translation. In this work, we extract parallel paragraphs from Paracrawl
parallel webpages using automatic sentence alignments and we use the extracted
parallel paragraphs as parallel documents for training document-level
translation models. We show that document-level NMT models trained with only
parallel paragraphs from Paracrawl can be used to translate real documents from
TED, News and Europarl, outperforming sentence-level NMT models. We also
perform a targeted pronoun evaluation and show that document-level models
trained with Paracrawl data can help context-aware pronoun translation.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 11:21:34 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Ghussin",
"Yusser Al",
""
],
[
"Zhang",
"Jingyi",
""
],
[
"van Genabith",
"Josef",
""
]
] | TITLE: Exploring Paracrawl for Document-level Neural Machine Translation
ABSTRACT: Document-level neural machine translation (NMT) has outperformed
sentence-level NMT on a number of datasets. However, document-level NMT is
still not widely adopted in real-world translation systems mainly due to the
lack of large-scale general-domain training data for document-level NMT. We
examine the effectiveness of using Paracrawl for learning document-level
translation. Paracrawl is a large-scale parallel corpus crawled from the
Internet and contains data from various domains. The official Paracrawl corpus
was released as parallel sentences (extracted from parallel webpages) and
therefore previous works only used Paracrawl for learning sentence-level
translation. In this work, we extract parallel paragraphs from Paracrawl
parallel webpages using automatic sentence alignments and we use the extracted
parallel paragraphs as parallel documents for training document-level
translation models. We show that document-level NMT models trained with only
parallel paragraphs from Paracrawl can be used to translate real documents from
TED, News and Europarl, outperforming sentence-level NMT models. We also
perform a targeted pronoun evaluation and show that document-level models
trained with Paracrawl data can help context-aware pronoun translation.
|
2304.10220 | XiaoKang Liu | Xiaokang Liu, Jianquan Li, Jingjing Mu, Min Yang, Ruifeng Xu, and
Benyou Wang | Effective Open Intent Classification with K-center Contrastive Learning
and Adjustable Decision Boundary | 9 pages, 4 figures | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Open intent classification, which aims to correctly classify the known
intents into their corresponding classes while identifying the new unknown
(open) intents, is an essential but challenging task in dialogue systems. In
this paper, we introduce novel K-center contrastive learning and adjustable
decision boundary learning (CLAB) to improve the effectiveness of open intent
classification. First, we pre-train a feature encoder on the labeled training
instances, which transfers knowledge from known intents to unknown intents.
Specifically, we devise a K-center contrastive learning algorithm to learn
discriminative and balanced intent features, improving the generalization of
the model for recognizing open intents. Second, we devise an adjustable
decision boundary learning method with expanding and shrinking (ADBES) to
determine the suitable decision conditions. Concretely, we learn a decision
boundary for each known intent class, which consists of a decision center and
the radius of the decision boundary. We then expand the radius of the decision
boundary to accommodate more in-class instances if the out-of-class instances
are far from the decision boundary; otherwise, we shrink the radius of the
decision boundary. Extensive experiments on three benchmark datasets clearly
demonstrate the effectiveness of our method for open intent classification. For
reproducibility, we submit the code at: https://github.com/lxk00/CLAP
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 11:35:06 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Liu",
"Xiaokang",
""
],
[
"Li",
"Jianquan",
""
],
[
"Mu",
"Jingjing",
""
],
[
"Yang",
"Min",
""
],
[
"Xu",
"Ruifeng",
""
],
[
"Wang",
"Benyou",
""
]
] | TITLE: Effective Open Intent Classification with K-center Contrastive Learning
and Adjustable Decision Boundary
ABSTRACT: Open intent classification, which aims to correctly classify the known
intents into their corresponding classes while identifying the new unknown
(open) intents, is an essential but challenging task in dialogue systems. In
this paper, we introduce novel K-center contrastive learning and adjustable
decision boundary learning (CLAB) to improve the effectiveness of open intent
classification. First, we pre-train a feature encoder on the labeled training
instances, which transfers knowledge from known intents to unknown intents.
Specifically, we devise a K-center contrastive learning algorithm to learn
discriminative and balanced intent features, improving the generalization of
the model for recognizing open intents. Second, we devise an adjustable
decision boundary learning method with expanding and shrinking (ADBES) to
determine the suitable decision conditions. Concretely, we learn a decision
boundary for each known intent class, which consists of a decision center and
the radius of the decision boundary. We then expand the radius of the decision
boundary to accommodate more in-class instances if the out-of-class instances
are far from the decision boundary; otherwise, we shrink the radius of the
decision boundary. Extensive experiments on three benchmark datasets clearly
demonstrate the effectiveness of our method for open intent classification. For
reproducibility, we submit the code at: https://github.com/lxk00/CLAP
|
2304.10241 | Yongming Yang | Yongming Yang, Shuwei Shao, Tao Yang, Peng Wang, Zhuo Yang, Chengdong
Wu, Hao Liu | A geometry-aware deep network for depth estimation in monocular
endoscopy | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Monocular depth estimation is critical for endoscopists to perform spatial
perception and 3D navigation of surgical sites. However, most of the existing
methods ignore the important geometric structural consistency, which inevitably
leads to performance degradation and distortion of 3D reconstruction. To
address this issue, we introduce a gradient loss to penalize edge fluctuations
ambiguous around stepped edge structures and a normal loss to explicitly
express the sensitivity to frequently small structures, and propose a geometric
consistency loss to spreads the spatial information across the sample grids to
constrain the global geometric anatomy structures. In addition, we develop a
synthetic RGB-Depth dataset that captures the anatomical structures under
reflections and illumination variations. The proposed method is extensively
validated across different datasets and clinical images and achieves mean RMSE
values of 0.066 (stomach), 0.029 (small intestine), and 0.139 (colon) on the
EndoSLAM dataset. The generalizability of the proposed method achieves mean
RMSE values of 12.604 (T1-L1), 9.930 (T2-L2), and 13.893 (T3-L3) on the
ColonDepth dataset. The experimental results show that our method exceeds
previous state-of-the-art competitors and generates more consistent depth maps
and reasonable anatomical structures. The quality of intraoperative 3D
structure perception from endoscopic videos of the proposed method meets the
accuracy requirements of video-CT registration algorithms for endoscopic
navigation. The dataset and the source code will be available at
https://github.com/YYM-SIA/LINGMI-MR.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 11:59:32 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Yang",
"Yongming",
""
],
[
"Shao",
"Shuwei",
""
],
[
"Yang",
"Tao",
""
],
[
"Wang",
"Peng",
""
],
[
"Yang",
"Zhuo",
""
],
[
"Wu",
"Chengdong",
""
],
[
"Liu",
"Hao",
""
]
] | TITLE: A geometry-aware deep network for depth estimation in monocular
endoscopy
ABSTRACT: Monocular depth estimation is critical for endoscopists to perform spatial
perception and 3D navigation of surgical sites. However, most of the existing
methods ignore the important geometric structural consistency, which inevitably
leads to performance degradation and distortion of 3D reconstruction. To
address this issue, we introduce a gradient loss to penalize edge fluctuations
ambiguous around stepped edge structures and a normal loss to explicitly
express the sensitivity to frequently small structures, and propose a geometric
consistency loss to spreads the spatial information across the sample grids to
constrain the global geometric anatomy structures. In addition, we develop a
synthetic RGB-Depth dataset that captures the anatomical structures under
reflections and illumination variations. The proposed method is extensively
validated across different datasets and clinical images and achieves mean RMSE
values of 0.066 (stomach), 0.029 (small intestine), and 0.139 (colon) on the
EndoSLAM dataset. The generalizability of the proposed method achieves mean
RMSE values of 12.604 (T1-L1), 9.930 (T2-L2), and 13.893 (T3-L3) on the
ColonDepth dataset. The experimental results show that our method exceeds
previous state-of-the-art competitors and generates more consistent depth maps
and reasonable anatomical structures. The quality of intraoperative 3D
structure perception from endoscopic videos of the proposed method meets the
accuracy requirements of video-CT registration algorithms for endoscopic
navigation. The dataset and the source code will be available at
https://github.com/YYM-SIA/LINGMI-MR.
|
2304.10261 | Qiuhong Shen | Qiuhong Shen, Xingyi Yang, Xinchao Wang | Anything-3D: Towards Single-view Anything Reconstruction in the Wild | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | 3D reconstruction from a single-RGB image in unconstrained real-world
scenarios presents numerous challenges due to the inherent diversity and
complexity of objects and environments. In this paper, we introduce
Anything-3D, a methodical framework that ingeniously combines a series of
visual-language models and the Segment-Anything object segmentation model to
elevate objects to 3D, yielding a reliable and versatile system for single-view
conditioned 3D reconstruction task. Our approach employs a BLIP model to
generate textural descriptions, utilizes the Segment-Anything model for the
effective extraction of objects of interest, and leverages a text-to-image
diffusion model to lift object into a neural radiance field. Demonstrating its
ability to produce accurate and detailed 3D reconstructions for a wide array of
objects, \emph{Anything-3D\footnotemark[2]} shows promise in addressing the
limitations of existing methodologies. Through comprehensive experiments and
evaluations on various datasets, we showcase the merits of our approach,
underscoring its potential to contribute meaningfully to the field of 3D
reconstruction. Demos and code will be available at
\href{https://github.com/Anything-of-anything/Anything-3D}{https://github.com/Anything-of-anything/Anything-3D}.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 16:39:51 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Shen",
"Qiuhong",
""
],
[
"Yang",
"Xingyi",
""
],
[
"Wang",
"Xinchao",
""
]
] | TITLE: Anything-3D: Towards Single-view Anything Reconstruction in the Wild
ABSTRACT: 3D reconstruction from a single-RGB image in unconstrained real-world
scenarios presents numerous challenges due to the inherent diversity and
complexity of objects and environments. In this paper, we introduce
Anything-3D, a methodical framework that ingeniously combines a series of
visual-language models and the Segment-Anything object segmentation model to
elevate objects to 3D, yielding a reliable and versatile system for single-view
conditioned 3D reconstruction task. Our approach employs a BLIP model to
generate textural descriptions, utilizes the Segment-Anything model for the
effective extraction of objects of interest, and leverages a text-to-image
diffusion model to lift object into a neural radiance field. Demonstrating its
ability to produce accurate and detailed 3D reconstructions for a wide array of
objects, \emph{Anything-3D\footnotemark[2]} shows promise in addressing the
limitations of existing methodologies. Through comprehensive experiments and
evaluations on various datasets, we showcase the merits of our approach,
underscoring its potential to contribute meaningfully to the field of 3D
reconstruction. Demos and code will be available at
\href{https://github.com/Anything-of-anything/Anything-3D}{https://github.com/Anything-of-anything/Anything-3D}.
|
2304.10283 | Marcos Prates O | Gabriel O. Assun\c{c}\~ao and Rafael Izbicki and Marcos O. Prates | Is augmentation effective to improve prediction in imbalanced text
datasets? | 21 pages, 5 figures | null | null | null | cs.CL stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Imbalanced datasets present a significant challenge for machine learning
models, often leading to biased predictions. To address this issue, data
augmentation techniques are widely used in natural language processing (NLP) to
generate new samples for the minority class. However, in this paper, we
challenge the common assumption that data augmentation is always necessary to
improve predictions on imbalanced datasets. Instead, we argue that adjusting
the classifier cutoffs without data augmentation can produce similar results to
oversampling techniques. Our study provides theoretical and empirical evidence
to support this claim. Our findings contribute to a better understanding of the
strengths and limitations of different approaches to dealing with imbalanced
data, and help researchers and practitioners make informed decisions about
which methods to use for a given task.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 13:07:31 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Assunção",
"Gabriel O.",
""
],
[
"Izbicki",
"Rafael",
""
],
[
"Prates",
"Marcos O.",
""
]
] | TITLE: Is augmentation effective to improve prediction in imbalanced text
datasets?
ABSTRACT: Imbalanced datasets present a significant challenge for machine learning
models, often leading to biased predictions. To address this issue, data
augmentation techniques are widely used in natural language processing (NLP) to
generate new samples for the minority class. However, in this paper, we
challenge the common assumption that data augmentation is always necessary to
improve predictions on imbalanced datasets. Instead, we argue that adjusting
the classifier cutoffs without data augmentation can produce similar results to
oversampling techniques. Our study provides theoretical and empirical evidence
to support this claim. Our findings contribute to a better understanding of the
strengths and limitations of different approaches to dealing with imbalanced
data, and help researchers and practitioners make informed decisions about
which methods to use for a given task.
|
2304.10309 | Hao Zhang | Hao Zhang, Nianwen Si, Yaqi Chen, Wenlin Zhang, Xukui Yang, Dan Qu,
and Wei-Qiang Zhang | Improving Speech Translation by Cross-Modal Multi-Grained Contrastive
Learning | null | IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING,
VOL. 31, 2023 | 10.1109/TASLP.2023.3244521 | null | cs.CL cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The end-to-end speech translation (E2E-ST) model has gradually become a
mainstream paradigm due to its low latency and less error propagation. However,
it is non-trivial to train such a model well due to the task complexity and
data scarcity. The speech-and-text modality differences result in the E2E-ST
model performance usually inferior to the corresponding machine translation
(MT) model. Based on the above observation, existing methods often use
sharingmechanisms to carry out implicit knowledge transfer by imposing various
constraints. However, the final model often performs worse on the MT task than
the MT model trained alone, which means that the knowledge transfer ability of
this method is also limited. To deal with these problems, we propose the FCCL
(Fine- and Coarse- Granularity Contrastive Learning) approach for E2E-ST, which
makes explicit knowledge transfer through cross-modal multi-grained contrastive
learning. A key ingredient of our approach is applying contrastive learning at
both sentence- and frame-level to give the comprehensive guide for extracting
speech representations containing rich semantic information.In addition, we
adopt a simple whitening method to alleviate the representation degeneration in
the MT model, which adversely affects contrast learning. Experiments on the
MuST-C benchmark show that our proposed approach significantly outperforms the
state-of-the-art E2E-ST baselines on all eight language pairs. Further analysis
indicates that FCCL can free up its capacity from learning grammatical
structure information and force more layers to learn semantic information.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 13:41:56 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Zhang",
"Hao",
""
],
[
"Si",
"Nianwen",
""
],
[
"Chen",
"Yaqi",
""
],
[
"Zhang",
"Wenlin",
""
],
[
"Yang",
"Xukui",
""
],
[
"Qu",
"Dan",
""
],
[
"Zhang",
"Wei-Qiang",
""
]
] | TITLE: Improving Speech Translation by Cross-Modal Multi-Grained Contrastive
Learning
ABSTRACT: The end-to-end speech translation (E2E-ST) model has gradually become a
mainstream paradigm due to its low latency and less error propagation. However,
it is non-trivial to train such a model well due to the task complexity and
data scarcity. The speech-and-text modality differences result in the E2E-ST
model performance usually inferior to the corresponding machine translation
(MT) model. Based on the above observation, existing methods often use
sharingmechanisms to carry out implicit knowledge transfer by imposing various
constraints. However, the final model often performs worse on the MT task than
the MT model trained alone, which means that the knowledge transfer ability of
this method is also limited. To deal with these problems, we propose the FCCL
(Fine- and Coarse- Granularity Contrastive Learning) approach for E2E-ST, which
makes explicit knowledge transfer through cross-modal multi-grained contrastive
learning. A key ingredient of our approach is applying contrastive learning at
both sentence- and frame-level to give the comprehensive guide for extracting
speech representations containing rich semantic information.In addition, we
adopt a simple whitening method to alleviate the representation degeneration in
the MT model, which adversely affects contrast learning. Experiments on the
MuST-C benchmark show that our proposed approach significantly outperforms the
state-of-the-art E2E-ST baselines on all eight language pairs. Further analysis
indicates that FCCL can free up its capacity from learning grammatical
structure information and force more layers to learn semantic information.
|
2304.10310 | Mingjun Zhao | Mingjun Zhao, Shan Lu, Zixuan Wang, Xiaoli Wang and Di Niu | LA3: Efficient Label-Aware AutoAugment | ECCV 2022 | null | 10.1007/978-3-031-19803-8_16 | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Automated augmentation is an emerging and effective technique to search for
data augmentation policies to improve generalizability of deep neural network
training. Most existing work focuses on constructing a unified policy
applicable to all data samples in a given dataset, without considering sample
or class variations. In this paper, we propose a novel two-stage data
augmentation algorithm, named Label-Aware AutoAugment (LA3), which takes
advantage of the label information, and learns augmentation policies separately
for samples of different labels. LA3 consists of two learning stages, where in
the first stage, individual augmentation methods are evaluated and ranked for
each label via Bayesian Optimization aided by a neural predictor, which allows
us to identify effective augmentation techniques for each label under a low
search cost. And in the second stage, a composite augmentation policy is
constructed out of a selection of effective as well as complementary
augmentations, which produces significant performance boost and can be easily
deployed in typical model training. Extensive experiments demonstrate that LA3
achieves excellent performance matching or surpassing existing methods on
CIFAR-10 and CIFAR-100, and achieves a new state-of-the-art ImageNet accuracy
of 79.97% on ResNet-50 among auto-augmentation methods, while maintaining a low
computational cost.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 13:42:18 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Zhao",
"Mingjun",
""
],
[
"Lu",
"Shan",
""
],
[
"Wang",
"Zixuan",
""
],
[
"Wang",
"Xiaoli",
""
],
[
"Niu",
"Di",
""
]
] | TITLE: LA3: Efficient Label-Aware AutoAugment
ABSTRACT: Automated augmentation is an emerging and effective technique to search for
data augmentation policies to improve generalizability of deep neural network
training. Most existing work focuses on constructing a unified policy
applicable to all data samples in a given dataset, without considering sample
or class variations. In this paper, we propose a novel two-stage data
augmentation algorithm, named Label-Aware AutoAugment (LA3), which takes
advantage of the label information, and learns augmentation policies separately
for samples of different labels. LA3 consists of two learning stages, where in
the first stage, individual augmentation methods are evaluated and ranked for
each label via Bayesian Optimization aided by a neural predictor, which allows
us to identify effective augmentation techniques for each label under a low
search cost. And in the second stage, a composite augmentation policy is
constructed out of a selection of effective as well as complementary
augmentations, which produces significant performance boost and can be easily
deployed in typical model training. Extensive experiments demonstrate that LA3
achieves excellent performance matching or surpassing existing methods on
CIFAR-10 and CIFAR-100, and achieves a new state-of-the-art ImageNet accuracy
of 79.97% on ResNet-50 among auto-augmentation methods, while maintaining a low
computational cost.
|
2304.10311 | Qin Chao | Qin Chao, Eunsoo Kim, Boyang Li | Movie Box Office Prediction With Self-Supervised and Visually Grounded
Pretraining | accepted by IEEE International Conference on Multimedia and Expo
(2023) | null | null | null | cs.MM cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Investments in movie production are associated with a high level of risk as
movie revenues have long-tailed and bimodal distributions. Accurate prediction
of box-office revenue may mitigate the uncertainty and encourage investment.
However, learning effective representations for actors, directors, and
user-generated content-related keywords remains a challenging open problem. In
this work, we investigate the effects of self-supervised pretraining and
propose visual grounding of content keywords in objects from movie posters as a
pertaining objective. Experiments on a large dataset of 35,794 movies
demonstrate significant benefits of self-supervised training and visual
grounding. In particular, visual grounding pretraining substantially improves
learning on movies with content keywords and achieves 14.5% relative
performance gains compared to a finetuned BERT model with identical
architecture.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 13:42:27 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Chao",
"Qin",
""
],
[
"Kim",
"Eunsoo",
""
],
[
"Li",
"Boyang",
""
]
] | TITLE: Movie Box Office Prediction With Self-Supervised and Visually Grounded
Pretraining
ABSTRACT: Investments in movie production are associated with a high level of risk as
movie revenues have long-tailed and bimodal distributions. Accurate prediction
of box-office revenue may mitigate the uncertainty and encourage investment.
However, learning effective representations for actors, directors, and
user-generated content-related keywords remains a challenging open problem. In
this work, we investigate the effects of self-supervised pretraining and
propose visual grounding of content keywords in objects from movie posters as a
pertaining objective. Experiments on a large dataset of 35,794 movies
demonstrate significant benefits of self-supervised training and visual
grounding. In particular, visual grounding pretraining substantially improves
learning on movies with content keywords and achieves 14.5% relative
performance gains compared to a finetuned BERT model with identical
architecture.
|
2304.10316 | Mingjun Zhao | Mingjun Zhao, Yakun Yu, Xiaoli Wang, Lei Yang and Di Niu | Search-Map-Search: A Frame Selection Paradigm for Action Recognition | CVPR 2023 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Despite the success of deep learning in video understanding tasks, processing
every frame in a video is computationally expensive and often unnecessary in
real-time applications. Frame selection aims to extract the most informative
and representative frames to help a model better understand video content.
Existing frame selection methods either individually sample frames based on
per-frame importance prediction, without considering interaction among frames,
or adopt reinforcement learning agents to find representative frames in
succession, which are costly to train and may lead to potential stability
issues. To overcome the limitations of existing methods, we propose a
Search-Map-Search learning paradigm which combines the advantages of heuristic
search and supervised learning to select the best combination of frames from a
video as one entity. By combining search with learning, the proposed method can
better capture frame interactions while incurring a low inference overhead.
Specifically, we first propose a hierarchical search method conducted on each
training video to search for the optimal combination of frames with the lowest
error on the downstream task. A feature mapping function is then learned to map
the frames of a video to the representation of its target optimal frame
combination. During inference, another search is performed on an unseen video
to select a combination of frames whose feature representation is close to the
projected feature representation. Extensive experiments based on several action
recognition benchmarks demonstrate that our frame selection method effectively
improves performance of action recognition models, and significantly
outperforms a number of competitive baselines.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 13:49:53 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Zhao",
"Mingjun",
""
],
[
"Yu",
"Yakun",
""
],
[
"Wang",
"Xiaoli",
""
],
[
"Yang",
"Lei",
""
],
[
"Niu",
"Di",
""
]
] | TITLE: Search-Map-Search: A Frame Selection Paradigm for Action Recognition
ABSTRACT: Despite the success of deep learning in video understanding tasks, processing
every frame in a video is computationally expensive and often unnecessary in
real-time applications. Frame selection aims to extract the most informative
and representative frames to help a model better understand video content.
Existing frame selection methods either individually sample frames based on
per-frame importance prediction, without considering interaction among frames,
or adopt reinforcement learning agents to find representative frames in
succession, which are costly to train and may lead to potential stability
issues. To overcome the limitations of existing methods, we propose a
Search-Map-Search learning paradigm which combines the advantages of heuristic
search and supervised learning to select the best combination of frames from a
video as one entity. By combining search with learning, the proposed method can
better capture frame interactions while incurring a low inference overhead.
Specifically, we first propose a hierarchical search method conducted on each
training video to search for the optimal combination of frames with the lowest
error on the downstream task. A feature mapping function is then learned to map
the frames of a video to the representation of its target optimal frame
combination. During inference, another search is performed on an unseen video
to select a combination of frames whose feature representation is close to the
projected feature representation. Extensive experiments based on several action
recognition benchmarks demonstrate that our frame selection method effectively
improves performance of action recognition models, and significantly
outperforms a number of competitive baselines.
|
2304.10317 | Pawan Kumar | Sachin Kumar Danisetty, Santhosh Reddy Mylaram, Pawan Kumar | Adaptive Consensus Optimization Method for GANs | null | null | null | null | cs.LG cs.CV cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | We propose a second order gradient based method with ADAM and RMSprop for the
training of generative adversarial networks. The proposed method is fastest to
obtain similar accuracy when compared to prominent second order methods. Unlike
state-of-the-art recent methods, it does not require solving a linear system,
or it does not require additional mixed second derivative terms. We derive the
fixed point iteration corresponding to proposed method, and show that the
proposed method is convergent. The proposed method produces better or
comparable inception scores, and comparable quality of images compared to other
recently proposed state-of-the-art second order methods. Compared to first
order methods such as ADAM, it produces significantly better inception scores.
The proposed method is compared and validated on popular datasets such as FFHQ,
LSUN, CIFAR10, MNIST, and Fashion MNIST for image generation
tasks\footnote{Accepted in IJCNN 2023}. Codes:
\url{https://github.com/misterpawan/acom}
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 13:50:42 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Danisetty",
"Sachin Kumar",
""
],
[
"Mylaram",
"Santhosh Reddy",
""
],
[
"Kumar",
"Pawan",
""
]
] | TITLE: Adaptive Consensus Optimization Method for GANs
ABSTRACT: We propose a second order gradient based method with ADAM and RMSprop for the
training of generative adversarial networks. The proposed method is fastest to
obtain similar accuracy when compared to prominent second order methods. Unlike
state-of-the-art recent methods, it does not require solving a linear system,
or it does not require additional mixed second derivative terms. We derive the
fixed point iteration corresponding to proposed method, and show that the
proposed method is convergent. The proposed method produces better or
comparable inception scores, and comparable quality of images compared to other
recently proposed state-of-the-art second order methods. Compared to first
order methods such as ADAM, it produces significantly better inception scores.
The proposed method is compared and validated on popular datasets such as FFHQ,
LSUN, CIFAR10, MNIST, and Fashion MNIST for image generation
tasks\footnote{Accepted in IJCNN 2023}. Codes:
\url{https://github.com/misterpawan/acom}
|
2304.10319 | Joseph Lee | Joseph K. L. Lee, Maurice Jamieson, Nick Brown, Ricardo Jesus | Test-driving RISC-V Vector hardware for HPC | Preprint of paper accepted to First International Workshop on RISC-V
for HPC (2023) | null | null | null | cs.DC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Whilst the RISC-V Vector extension (RVV) has been ratified, at the time of
writing both hardware implementations and open source software support are
still limited for vectorisation on RISC-V. This is important because
vectorisation is crucial to obtaining good performance for High Performance
Computing (HPC) workloads and, as of April 2023, the Allwinner D1 SoC,
containing the XuanTie C906 processor, is the only mass-produced and
commercially available hardware supporting RVV. This paper surveys the current
state of RISC-V vectorisation as of 2023, reporting the landscape of both the
hardware and software ecosystem. Driving our discussion from experiences in
setting up the Allwinner D1 as part of the EPCC RISC-V testbed, we report the
results of benchmarking the Allwinner D1 using the RAJA Performance Suite,
which demonstrated reasonable vectorisation speedup using vendor-provided
compiler, as well as favourable performance compared to the StarFive VisionFive
V2 with SiFive's U74 processor.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 13:51:42 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Lee",
"Joseph K. L.",
""
],
[
"Jamieson",
"Maurice",
""
],
[
"Brown",
"Nick",
""
],
[
"Jesus",
"Ricardo",
""
]
] | TITLE: Test-driving RISC-V Vector hardware for HPC
ABSTRACT: Whilst the RISC-V Vector extension (RVV) has been ratified, at the time of
writing both hardware implementations and open source software support are
still limited for vectorisation on RISC-V. This is important because
vectorisation is crucial to obtaining good performance for High Performance
Computing (HPC) workloads and, as of April 2023, the Allwinner D1 SoC,
containing the XuanTie C906 processor, is the only mass-produced and
commercially available hardware supporting RVV. This paper surveys the current
state of RISC-V vectorisation as of 2023, reporting the landscape of both the
hardware and software ecosystem. Driving our discussion from experiences in
setting up the Allwinner D1 as part of the EPCC RISC-V testbed, we report the
results of benchmarking the Allwinner D1 using the RAJA Performance Suite,
which demonstrated reasonable vectorisation speedup using vendor-provided
compiler, as well as favourable performance compared to the StarFive VisionFive
V2 with SiFive's U74 processor.
|
2304.10321 | Hao Zhang | Hao Zhang, Dan Qu, Keji Shao, and Xukui Yang | DropDim: A Regularization Method for Transformer Networks | null | IEEE SIGNAL PROCESSING LETTERS, VOL. 29, 2022 | 10.1109/LSP.2022.3140693 | null | cs.CL cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduceDropDim, a structured dropout method designed for regularizing
the self-attention mechanism, which is a key component of the transformer. In
contrast to the general dropout method, which randomly drops neurons, DropDim
drops part of the embedding dimensions. In this way, the semantic information
can be completely discarded. Thus, the excessive coadapting between different
embedding dimensions can be broken, and the self-attention is forced to encode
meaningful featureswith a certain number of embedding dimensions erased.
Experiments on a wide range of tasks executed on the MUST-C English-Germany
dataset show that DropDim can effectively improve model performance, reduce
over-fitting, and show complementary effects with other regularization methods.
When combined with label smoothing, the WER can be reduced from 19.1% to 15.1%
on the ASR task, and the BLEU value can be increased from26.90 to 28.38 on the
MT task. On the ST task, the model can reach a BLEU score of 22.99, an increase
by 1.86 BLEU points compared to the strong baseline.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 13:54:18 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Zhang",
"Hao",
""
],
[
"Qu",
"Dan",
""
],
[
"Shao",
"Keji",
""
],
[
"Yang",
"Xukui",
""
]
] | TITLE: DropDim: A Regularization Method for Transformer Networks
ABSTRACT: We introduceDropDim, a structured dropout method designed for regularizing
the self-attention mechanism, which is a key component of the transformer. In
contrast to the general dropout method, which randomly drops neurons, DropDim
drops part of the embedding dimensions. In this way, the semantic information
can be completely discarded. Thus, the excessive coadapting between different
embedding dimensions can be broken, and the self-attention is forced to encode
meaningful featureswith a certain number of embedding dimensions erased.
Experiments on a wide range of tasks executed on the MUST-C English-Germany
dataset show that DropDim can effectively improve model performance, reduce
over-fitting, and show complementary effects with other regularization methods.
When combined with label smoothing, the WER can be reduced from 19.1% to 15.1%
on the ASR task, and the BLEU value can be increased from26.90 to 28.38 on the
MT task. On the ST task, the model can reach a BLEU score of 22.99, an increase
by 1.86 BLEU points compared to the strong baseline.
|
2304.10354 | Liang Ding | Chiaming Hsu, Changtong Zan, Liang Ding, Longyue Wang, Xiaoting Wang,
Weifeng Liu, Fu Lin, Wenbin Hu | Prompt-Learning for Cross-Lingual Relation Extraction | IJCNN 2023 | null | null | null | cs.CL | http://creativecommons.org/publicdomain/zero/1.0/ | Relation Extraction (RE) is a crucial task in Information Extraction, which
entails predicting relationships between entities within a given sentence.
However, extending pre-trained RE models to other languages is challenging,
particularly in real-world scenarios where Cross-Lingual Relation Extraction
(XRE) is required. Despite recent advancements in Prompt-Learning, which
involves transferring knowledge from Multilingual Pre-trained Language Models
(PLMs) to diverse downstream tasks, there is limited research on the effective
use of multilingual PLMs with prompts to improve XRE. In this paper, we present
a novel XRE algorithm based on Prompt-Tuning, referred to as Prompt-XRE. To
evaluate its effectiveness, we design and implement several prompt templates,
including hard, soft, and hybrid prompts, and empirically test their
performance on competitive multilingual PLMs, specifically mBART. Our extensive
experiments, conducted on the low-resource ACE05 benchmark across multiple
languages, demonstrate that our Prompt-XRE algorithm significantly outperforms
both vanilla multilingual PLMs and other existing models, achieving
state-of-the-art performance in XRE. To further show the generalization of our
Prompt-XRE on larger data scales, we construct and release a new XRE dataset-
WMT17-EnZh XRE, containing 0.9M English-Chinese pairs extracted from WMT 2017
parallel corpus. Experiments on WMT17-EnZh XRE also show the effectiveness of
our Prompt-XRE against other competitive baselines. The code and newly
constructed dataset are freely available at
\url{https://github.com/HSU-CHIA-MING/Prompt-XRE}.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 14:52:06 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Hsu",
"Chiaming",
""
],
[
"Zan",
"Changtong",
""
],
[
"Ding",
"Liang",
""
],
[
"Wang",
"Longyue",
""
],
[
"Wang",
"Xiaoting",
""
],
[
"Liu",
"Weifeng",
""
],
[
"Lin",
"Fu",
""
],
[
"Hu",
"Wenbin",
""
]
] | TITLE: Prompt-Learning for Cross-Lingual Relation Extraction
ABSTRACT: Relation Extraction (RE) is a crucial task in Information Extraction, which
entails predicting relationships between entities within a given sentence.
However, extending pre-trained RE models to other languages is challenging,
particularly in real-world scenarios where Cross-Lingual Relation Extraction
(XRE) is required. Despite recent advancements in Prompt-Learning, which
involves transferring knowledge from Multilingual Pre-trained Language Models
(PLMs) to diverse downstream tasks, there is limited research on the effective
use of multilingual PLMs with prompts to improve XRE. In this paper, we present
a novel XRE algorithm based on Prompt-Tuning, referred to as Prompt-XRE. To
evaluate its effectiveness, we design and implement several prompt templates,
including hard, soft, and hybrid prompts, and empirically test their
performance on competitive multilingual PLMs, specifically mBART. Our extensive
experiments, conducted on the low-resource ACE05 benchmark across multiple
languages, demonstrate that our Prompt-XRE algorithm significantly outperforms
both vanilla multilingual PLMs and other existing models, achieving
state-of-the-art performance in XRE. To further show the generalization of our
Prompt-XRE on larger data scales, we construct and release a new XRE dataset-
WMT17-EnZh XRE, containing 0.9M English-Chinese pairs extracted from WMT 2017
parallel corpus. Experiments on WMT17-EnZh XRE also show the effectiveness of
our Prompt-XRE against other competitive baselines. The code and newly
constructed dataset are freely available at
\url{https://github.com/HSU-CHIA-MING/Prompt-XRE}.
|
2304.10386 | Gayathri Girish Kumar | Gayathri Girish, Ponnathota Spandana, Badrish Vasu | Breast cancer detection using deep learning | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Objective: This paper proposes a deep learning model for breast cancer
detection from reconstructed images of microwave imaging scan data and aims to
improve the accuracy and efficiency of breast tumor detection, which could have
a significant impact on breast cancer diagnosis and treatment. Methods: Our
framework consists of different convolutional neural network (CNN)
architectures for feature extraction and a region-based CNN for tumor
detection. We use 7 different architectures: DenseNet201, ResNet50,
InceptionV3, InceptionResNetV3, MobileNetV2, NASNetMobile and NASNetLarge and
compare its performance to find the best architecture out of the seven. An
experimental dataset of MRI-derived breast phantoms was used. Results:
NASNetLarge is the best architecture which can be used for the CNN model with
accuracy of 88.41% and loss of 27.82%. Given that the model's AUC is 0.786, it
can be concluded that it is suitable for use in its present form, while it
could be improved upon and trained on other datasets that are comparable.
Impact: One of the main causes of death in women is breast cancer, and early
identification is essential for enhancing the results for patients. Due to its
non-invasiveness and capacity to produce high-resolution images, microwave
imaging is a potential tool for breast cancer screening. The complexity of
tumors makes it difficult to adequately detect them in microwave images. The
results of this research show that deep learning has a lot of potential for
breast cancer detection in microwave images
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 15:22:12 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Girish",
"Gayathri",
""
],
[
"Spandana",
"Ponnathota",
""
],
[
"Vasu",
"Badrish",
""
]
] | TITLE: Breast cancer detection using deep learning
ABSTRACT: Objective: This paper proposes a deep learning model for breast cancer
detection from reconstructed images of microwave imaging scan data and aims to
improve the accuracy and efficiency of breast tumor detection, which could have
a significant impact on breast cancer diagnosis and treatment. Methods: Our
framework consists of different convolutional neural network (CNN)
architectures for feature extraction and a region-based CNN for tumor
detection. We use 7 different architectures: DenseNet201, ResNet50,
InceptionV3, InceptionResNetV3, MobileNetV2, NASNetMobile and NASNetLarge and
compare its performance to find the best architecture out of the seven. An
experimental dataset of MRI-derived breast phantoms was used. Results:
NASNetLarge is the best architecture which can be used for the CNN model with
accuracy of 88.41% and loss of 27.82%. Given that the model's AUC is 0.786, it
can be concluded that it is suitable for use in its present form, while it
could be improved upon and trained on other datasets that are comparable.
Impact: One of the main causes of death in women is breast cancer, and early
identification is essential for enhancing the results for patients. Due to its
non-invasiveness and capacity to produce high-resolution images, microwave
imaging is a potential tool for breast cancer screening. The complexity of
tumors makes it difficult to adequately detect them in microwave images. The
results of this research show that deep learning has a lot of potential for
breast cancer detection in microwave images
|
2304.10397 | Noel Chalmers | Noel Chalmers and Jakub Kurzak and Damon McDougall and Paul T. Bauman | Optimizing High-Performance Linpack for Exascale Accelerated
Architectures | null | null | null | null | cs.DC cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | We detail the performance optimizations made in rocHPL, AMD's open-source
implementation of the High-Performance Linpack (HPL) benchmark targeting
accelerated node architectures designed for exascale systems such as the
Frontier supercomputer. The implementation leverages the high-throughput GPU
accelerators on the node via highly optimized linear algebra libraries, as well
as the entire CPU socket to perform latency-sensitive factorization phases. We
detail novel performance improvements such as a multi-threaded approach to
computing the panel factorization phase on the CPU, time-sharing of CPU cores
between processes on the node, as well as several optimizations which hide MPI
communication. We present some performance results of this implementation of
the HPL benchmark on a single node of the Frontier early access cluster at Oak
Ridge National Laboratory, as well as scaling to multiple nodes.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 15:33:55 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Chalmers",
"Noel",
""
],
[
"Kurzak",
"Jakub",
""
],
[
"McDougall",
"Damon",
""
],
[
"Bauman",
"Paul T.",
""
]
] | TITLE: Optimizing High-Performance Linpack for Exascale Accelerated
Architectures
ABSTRACT: We detail the performance optimizations made in rocHPL, AMD's open-source
implementation of the High-Performance Linpack (HPL) benchmark targeting
accelerated node architectures designed for exascale systems such as the
Frontier supercomputer. The implementation leverages the high-throughput GPU
accelerators on the node via highly optimized linear algebra libraries, as well
as the entire CPU socket to perform latency-sensitive factorization phases. We
detail novel performance improvements such as a multi-threaded approach to
computing the panel factorization phase on the CPU, time-sharing of CPU cores
between processes on the node, as well as several optimizations which hide MPI
communication. We present some performance results of this implementation of
the HPL benchmark on a single node of the Frontier early access cluster at Oak
Ridge National Laboratory, as well as scaling to multiple nodes.
|
2304.10415 | Zhengyu Liang | Yingqian Wang, Longguang Wang, Zhengyu Liang, Jungang Yang, Radu
Timofte, Yulan Guo | NTIRE 2023 Challenge on Light Field Image Super-Resolution: Dataset,
Methods and Results | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | In this report, we summarize the first NTIRE challenge on light field (LF)
image super-resolution (SR), which aims at super-resolving LF images under the
standard bicubic degradation with a magnification factor of 4. This challenge
develops a new LF dataset called NTIRE-2023 for validation and test, and
provides a toolbox called BasicLFSR to facilitate model development. Compared
with single image SR, the major challenge of LF image SR lies in how to exploit
complementary angular information from plenty of views with varying
disparities. In total, 148 participants have registered the challenge, and 11
teams have successfully submitted results with PSNR scores higher than the
baseline method LF-InterNet \cite{LF-InterNet}. These newly developed methods
have set new state-of-the-art in LF image SR, e.g., the winning method achieves
around 1 dB PSNR improvement over the existing state-of-the-art method DistgSSR
\cite{DistgLF}. We report the solutions proposed by the participants, and
summarize their common trends and useful tricks. We hope this challenge can
stimulate future research and inspire new ideas in LF image SR.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 15:59:31 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Wang",
"Yingqian",
""
],
[
"Wang",
"Longguang",
""
],
[
"Liang",
"Zhengyu",
""
],
[
"Yang",
"Jungang",
""
],
[
"Timofte",
"Radu",
""
],
[
"Guo",
"Yulan",
""
]
] | TITLE: NTIRE 2023 Challenge on Light Field Image Super-Resolution: Dataset,
Methods and Results
ABSTRACT: In this report, we summarize the first NTIRE challenge on light field (LF)
image super-resolution (SR), which aims at super-resolving LF images under the
standard bicubic degradation with a magnification factor of 4. This challenge
develops a new LF dataset called NTIRE-2023 for validation and test, and
provides a toolbox called BasicLFSR to facilitate model development. Compared
with single image SR, the major challenge of LF image SR lies in how to exploit
complementary angular information from plenty of views with varying
disparities. In total, 148 participants have registered the challenge, and 11
teams have successfully submitted results with PSNR scores higher than the
baseline method LF-InterNet \cite{LF-InterNet}. These newly developed methods
have set new state-of-the-art in LF image SR, e.g., the winning method achieves
around 1 dB PSNR improvement over the existing state-of-the-art method DistgSSR
\cite{DistgLF}. We report the solutions proposed by the participants, and
summarize their common trends and useful tricks. We hope this challenge can
stimulate future research and inspire new ideas in LF image SR.
|
2304.10436 | Hao Sun | Hao Sun, Zhexin Zhang, Jiawen Deng, Jiale Cheng, Minlie Huang | Safety Assessment of Chinese Large Language Models | Benchmark website: http://coai.cs.tsinghua.edu.cn/leaderboard/ ;
SafetyPrompts repo: https://github.com/thu-coai/Safety-Prompts | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the rapid popularity of large language models such as ChatGPT and GPT-4,
a growing amount of attention is paid to their safety concerns. These models
may generate insulting and discriminatory content, reflect incorrect social
values, and may be used for malicious purposes such as fraud and dissemination
of misleading information. Evaluating and enhancing their safety is
particularly essential for the wide application of large language models
(LLMs). To further promote the safe deployment of LLMs, we develop a Chinese
LLM safety assessment benchmark. Our benchmark explores the comprehensive
safety performance of LLMs from two perspectives: 8 kinds of typical safety
scenarios and 6 types of more challenging instruction attacks. Our benchmark is
based on a straightforward process in which it provides the test prompts and
evaluates the safety of the generated responses from the evaluated model. In
evaluation, we utilize the LLM's strong evaluation ability and develop it as a
safety evaluator by prompting. On top of this benchmark, we conduct safety
assessments and analyze 15 LLMs including the OpenAI GPT series and other
well-known Chinese LLMs, where we observe some interesting findings. For
example, we find that instruction attacks are more likely to expose safety
issues of all LLMs. Moreover, to promote the development and deployment of
safe, responsible, and ethical AI, we publicly release SafetyPrompts including
100k augmented prompts and responses by LLMs.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 16:27:35 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Sun",
"Hao",
""
],
[
"Zhang",
"Zhexin",
""
],
[
"Deng",
"Jiawen",
""
],
[
"Cheng",
"Jiale",
""
],
[
"Huang",
"Minlie",
""
]
] | TITLE: Safety Assessment of Chinese Large Language Models
ABSTRACT: With the rapid popularity of large language models such as ChatGPT and GPT-4,
a growing amount of attention is paid to their safety concerns. These models
may generate insulting and discriminatory content, reflect incorrect social
values, and may be used for malicious purposes such as fraud and dissemination
of misleading information. Evaluating and enhancing their safety is
particularly essential for the wide application of large language models
(LLMs). To further promote the safe deployment of LLMs, we develop a Chinese
LLM safety assessment benchmark. Our benchmark explores the comprehensive
safety performance of LLMs from two perspectives: 8 kinds of typical safety
scenarios and 6 types of more challenging instruction attacks. Our benchmark is
based on a straightforward process in which it provides the test prompts and
evaluates the safety of the generated responses from the evaluated model. In
evaluation, we utilize the LLM's strong evaluation ability and develop it as a
safety evaluator by prompting. On top of this benchmark, we conduct safety
assessments and analyze 15 LLMs including the OpenAI GPT series and other
well-known Chinese LLMs, where we observe some interesting findings. For
example, we find that instruction attacks are more likely to expose safety
issues of all LLMs. Moreover, to promote the development and deployment of
safe, responsible, and ethical AI, we publicly release SafetyPrompts including
100k augmented prompts and responses by LLMs.
|
2304.10439 | Kehinde Ajayi | Kehinde Ajayi, Muntabir Hasan Choudhury, Sarah Rajtmajer, and Jian Wu | A Study on Reproducibility and Replicability of Table Structure
Recognition Methods | 10 pages, 5 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Concerns about reproducibility in artificial intelligence (AI) have emerged,
as researchers have reported unsuccessful attempts to directly reproduce
published findings in the field. Replicability, the ability to affirm a finding
using the same procedures on new data, has not been well studied. In this
paper, we examine both reproducibility and replicability of a corpus of 16
papers on table structure recognition (TSR), an AI task aimed at identifying
cell locations of tables in digital documents. We attempt to reproduce
published results using codes and datasets provided by the original authors. We
then examine replicability using a dataset similar to the original as well as a
new dataset, GenTSR, consisting of 386 annotated tables extracted from
scientific papers. Out of 16 papers studied, we reproduce results consistent
with the original in only four. Two of the four papers are identified as
replicable using the similar dataset under certain IoU values. No paper is
identified as replicable using the new dataset. We offer observations on the
causes of irreproducibility and irreplicability. All code and data are
available on Codeocean at https://codeocean.com/capsule/6680116/tree.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 16:30:58 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Ajayi",
"Kehinde",
""
],
[
"Choudhury",
"Muntabir Hasan",
""
],
[
"Rajtmajer",
"Sarah",
""
],
[
"Wu",
"Jian",
""
]
] | TITLE: A Study on Reproducibility and Replicability of Table Structure
Recognition Methods
ABSTRACT: Concerns about reproducibility in artificial intelligence (AI) have emerged,
as researchers have reported unsuccessful attempts to directly reproduce
published findings in the field. Replicability, the ability to affirm a finding
using the same procedures on new data, has not been well studied. In this
paper, we examine both reproducibility and replicability of a corpus of 16
papers on table structure recognition (TSR), an AI task aimed at identifying
cell locations of tables in digital documents. We attempt to reproduce
published results using codes and datasets provided by the original authors. We
then examine replicability using a dataset similar to the original as well as a
new dataset, GenTSR, consisting of 386 annotated tables extracted from
scientific papers. Out of 16 papers studied, we reproduce results consistent
with the original in only four. Two of the four papers are identified as
replicable using the similar dataset under certain IoU values. No paper is
identified as replicable using the new dataset. We offer observations on the
causes of irreproducibility and irreplicability. All code and data are
available on Codeocean at https://codeocean.com/capsule/6680116/tree.
|
2304.10448 | Riccardo Spezialetti | Marco Toschi, Riccardo De Matteo, Riccardo Spezialetti, Daniele De
Gregorio, Luigi Di Stefano, Samuele Salti | ReLight My NeRF: A Dataset for Novel View Synthesis and Relighting of
Real World Objects | Accepted at CVPR 2023 as a highlight | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we focus on the problem of rendering novel views from a Neural
Radiance Field (NeRF) under unobserved light conditions. To this end, we
introduce a novel dataset, dubbed ReNe (Relighting NeRF), framing real world
objects under one-light-at-time (OLAT) conditions, annotated with accurate
ground-truth camera and light poses. Our acquisition pipeline leverages two
robotic arms holding, respectively, a camera and an omni-directional point-wise
light source. We release a total of 20 scenes depicting a variety of objects
with complex geometry and challenging materials. Each scene includes 2000
images, acquired from 50 different points of views under 40 different OLAT
conditions. By leveraging the dataset, we perform an ablation study on the
relighting capability of variants of the vanilla NeRF architecture and identify
a lightweight architecture that can render novel views of an object under novel
light conditions, which we use to establish a non-trivial baseline for the
dataset. Dataset and benchmark are available at
https://eyecan-ai.github.io/rene.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 16:43:58 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Toschi",
"Marco",
""
],
[
"De Matteo",
"Riccardo",
""
],
[
"Spezialetti",
"Riccardo",
""
],
[
"De Gregorio",
"Daniele",
""
],
[
"Di Stefano",
"Luigi",
""
],
[
"Salti",
"Samuele",
""
]
] | TITLE: ReLight My NeRF: A Dataset for Novel View Synthesis and Relighting of
Real World Objects
ABSTRACT: In this paper, we focus on the problem of rendering novel views from a Neural
Radiance Field (NeRF) under unobserved light conditions. To this end, we
introduce a novel dataset, dubbed ReNe (Relighting NeRF), framing real world
objects under one-light-at-time (OLAT) conditions, annotated with accurate
ground-truth camera and light poses. Our acquisition pipeline leverages two
robotic arms holding, respectively, a camera and an omni-directional point-wise
light source. We release a total of 20 scenes depicting a variety of objects
with complex geometry and challenging materials. Each scene includes 2000
images, acquired from 50 different points of views under 40 different OLAT
conditions. By leveraging the dataset, we perform an ablation study on the
relighting capability of variants of the vanilla NeRF architecture and identify
a lightweight architecture that can render novel views of an object under novel
light conditions, which we use to establish a non-trivial baseline for the
dataset. Dataset and benchmark are available at
https://eyecan-ai.github.io/rene.
|
2304.10457 | Pawan Kumar | Neel Mishra, Pawan Kumar | Angle based dynamic learning rate for gradient descent | 8 pages, 7 figures | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by/4.0/ | In our work, we propose a novel yet simple approach to obtain an adaptive
learning rate for gradient-based descent methods on classification tasks.
Instead of the traditional approach of selecting adaptive learning rates via
the decayed expectation of gradient-based terms, we use the angle between the
current gradient and the new gradient: this new gradient is computed from the
direction orthogonal to the current gradient, which further helps us in
determining a better adaptive learning rate based on angle history, thereby,
leading to relatively better accuracy compared to the existing state-of-the-art
optimizers. On a wide variety of benchmark datasets with prominent image
classification architectures such as ResNet, DenseNet, EfficientNet, and VGG,
we find that our method leads to the highest accuracy in most of the datasets.
Moreover, we prove that our method is convergent.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 16:55:56 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Mishra",
"Neel",
""
],
[
"Kumar",
"Pawan",
""
]
] | TITLE: Angle based dynamic learning rate for gradient descent
ABSTRACT: In our work, we propose a novel yet simple approach to obtain an adaptive
learning rate for gradient-based descent methods on classification tasks.
Instead of the traditional approach of selecting adaptive learning rates via
the decayed expectation of gradient-based terms, we use the angle between the
current gradient and the new gradient: this new gradient is computed from the
direction orthogonal to the current gradient, which further helps us in
determining a better adaptive learning rate based on angle history, thereby,
leading to relatively better accuracy compared to the existing state-of-the-art
optimizers. On a wide variety of benchmark datasets with prominent image
classification architectures such as ResNet, DenseNet, EfficientNet, and VGG,
we find that our method leads to the highest accuracy in most of the datasets.
Moreover, we prove that our method is convergent.
|
2304.10494 | Haotian Zhang | Haotian Zhang, Jintu Zhang, Huifeng Zhao, Dejun Jiang, Yafeng Deng | Infinite Physical Monkey: Do Deep Learning Methods Really Perform Better
in Conformation Generation? | null | null | null | null | q-bio.BM cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Conformation Generation is a fundamental problem in drug discovery and
cheminformatics. And organic molecule conformation generation, particularly in
vacuum and protein pocket environments, is most relevant to drug design.
Recently, with the development of geometric neural networks, the data-driven
schemes have been successfully applied in this field, both for molecular
conformation generation (in vacuum) and binding pose generation (in protein
pocket). The former beats the traditional ETKDG method, while the latter
achieves similar accuracy compared with the widely used molecular docking
software. Although these methods have shown promising results, some researchers
have recently questioned whether deep learning (DL) methods perform better in
molecular conformation generation via a parameter-free method. To our surprise,
what they have designed is some kind analogous to the famous infinite monkey
theorem, the monkeys that are even equipped with physics education. To discuss
the feasibility of their proving, we constructed a real infinite stochastic
monkey for molecular conformation generation, showing that even with a more
stochastic sampler for geometry generation, the coverage of the benchmark
QM-computed conformations are higher than those of most DL-based methods. By
extending their physical monkey algorithm for binding pose prediction, we also
discover that the successful docking rate also achieves near-best performance
among existing DL-based docking models. Thus, though their conclusions are
right, their proof process needs more concern.
| [
{
"version": "v1",
"created": "Wed, 8 Mar 2023 02:09:58 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Zhang",
"Haotian",
""
],
[
"Zhang",
"Jintu",
""
],
[
"Zhao",
"Huifeng",
""
],
[
"Jiang",
"Dejun",
""
],
[
"Deng",
"Yafeng",
""
]
] | TITLE: Infinite Physical Monkey: Do Deep Learning Methods Really Perform Better
in Conformation Generation?
ABSTRACT: Conformation Generation is a fundamental problem in drug discovery and
cheminformatics. And organic molecule conformation generation, particularly in
vacuum and protein pocket environments, is most relevant to drug design.
Recently, with the development of geometric neural networks, the data-driven
schemes have been successfully applied in this field, both for molecular
conformation generation (in vacuum) and binding pose generation (in protein
pocket). The former beats the traditional ETKDG method, while the latter
achieves similar accuracy compared with the widely used molecular docking
software. Although these methods have shown promising results, some researchers
have recently questioned whether deep learning (DL) methods perform better in
molecular conformation generation via a parameter-free method. To our surprise,
what they have designed is some kind analogous to the famous infinite monkey
theorem, the monkeys that are even equipped with physics education. To discuss
the feasibility of their proving, we constructed a real infinite stochastic
monkey for molecular conformation generation, showing that even with a more
stochastic sampler for geometry generation, the coverage of the benchmark
QM-computed conformations are higher than those of most DL-based methods. By
extending their physical monkey algorithm for binding pose prediction, we also
discover that the successful docking rate also achieves near-best performance
among existing DL-based docking models. Thus, though their conclusions are
right, their proof process needs more concern.
|
2304.10500 | Brando Miranda | Brando Miranda, Avi Shinnar, Vasily Pestun, and Barry Trager | Transformer Models for Type Inference in the Simply Typed Lambda
Calculus: A Case Study in Deep Learning for Code | 22 pages | null | null | null | cs.PL cs.AI cs.LG cs.LO cs.SC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite a growing body of work at the intersection of deep learning and
formal languages, there has been relatively little systematic exploration of
transformer models for reasoning about typed lambda calculi. This is an
interesting area of inquiry for two reasons. First, typed lambda calculi are
the lingua franc of programming languages. A set of heuristics that relate
various typed lambda calculi to effective neural architectures would provide a
systematic method for mapping language features (e.g., polymorphism, subtyping,
inheritance, etc.) to architecture choices. Second, transformer models are
widely used in deep learning architectures applied to code, but the design and
hyperparameter space for them is large and relatively unexplored in programming
language applications. Therefore, we suggest a benchmark that allows us to
explore exactly this through perhaps the simplest and most fundamental property
of a programming language: the relationship between terms and types.
Consequently, we begin this inquiry of transformer architectures for typed
lambda calculi by exploring the effect of transformer warm-up and optimizer
selection in the task of type inference: i.e., predicting the types of lambda
calculus terms using only transformers. We find that the optimization landscape
is difficult even in this simple setting. One particular experimental finding
is that optimization by Adafactor converges much faster compared to the
optimization by Adam and RAdam. We conjecture that such different performance
of optimizers might be related to the difficulties of generalization over
formally generated dataset.
| [
{
"version": "v1",
"created": "Wed, 15 Mar 2023 17:44:39 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Miranda",
"Brando",
""
],
[
"Shinnar",
"Avi",
""
],
[
"Pestun",
"Vasily",
""
],
[
"Trager",
"Barry",
""
]
] | TITLE: Transformer Models for Type Inference in the Simply Typed Lambda
Calculus: A Case Study in Deep Learning for Code
ABSTRACT: Despite a growing body of work at the intersection of deep learning and
formal languages, there has been relatively little systematic exploration of
transformer models for reasoning about typed lambda calculi. This is an
interesting area of inquiry for two reasons. First, typed lambda calculi are
the lingua franc of programming languages. A set of heuristics that relate
various typed lambda calculi to effective neural architectures would provide a
systematic method for mapping language features (e.g., polymorphism, subtyping,
inheritance, etc.) to architecture choices. Second, transformer models are
widely used in deep learning architectures applied to code, but the design and
hyperparameter space for them is large and relatively unexplored in programming
language applications. Therefore, we suggest a benchmark that allows us to
explore exactly this through perhaps the simplest and most fundamental property
of a programming language: the relationship between terms and types.
Consequently, we begin this inquiry of transformer architectures for typed
lambda calculi by exploring the effect of transformer warm-up and optimizer
selection in the task of type inference: i.e., predicting the types of lambda
calculus terms using only transformers. We find that the optimization landscape
is difficult even in this simple setting. One particular experimental finding
is that optimization by Adafactor converges much faster compared to the
optimization by Adam and RAdam. We conjecture that such different performance
of optimizers might be related to the difficulties of generalization over
formally generated dataset.
|
2304.10505 | Kastan Day | Kastan Day, Daniel Christl, Rohan Salvi, Pranav Sriram | Video Pre-trained Transformer: A Multimodal Mixture of Pre-trained
Experts | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by-sa/4.0/ | We present Video Pre-trained Transformer. VPT uses four SOTA encoder models
from prior work to convert a video into a sequence of compact embeddings. Our
backbone, based on a reference Flan-T5-11B architecture, learns a universal
representation of the video that is a non-linear sum of the encoder models. It
learns using an autoregressive causal language modeling loss by predicting the
words spoken in YouTube videos. Finally, we evaluate on standard downstream
benchmarks by training fully connected prediction heads for each task. To the
best of our knowledge, this is the first use of multiple frozen SOTA models as
encoders in an "embedding -> backbone -> prediction head" design pattern - all
others have trained their own joint encoder models. Additionally, we include
more modalities than the current SOTA, Merlot Reserve, by adding explicit Scene
Graph information. For these two reasons, we believe it could combine the
world's best open-source models to achieve SOTA performance. Initial
experiments demonstrate the model is learning appropriately, but more
experimentation and compute is necessary, and already in progress, to realize
our loftier goals. Alongside this work, we build on the YT-20M dataset,
reproducing it and adding 25,000 personally selected YouTube videos to its
corpus. All code and model checkpoints are open sourced under a standard MIT
license.
| [
{
"version": "v1",
"created": "Fri, 24 Mar 2023 17:18:40 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Day",
"Kastan",
""
],
[
"Christl",
"Daniel",
""
],
[
"Salvi",
"Rohan",
""
],
[
"Sriram",
"Pranav",
""
]
] | TITLE: Video Pre-trained Transformer: A Multimodal Mixture of Pre-trained
Experts
ABSTRACT: We present Video Pre-trained Transformer. VPT uses four SOTA encoder models
from prior work to convert a video into a sequence of compact embeddings. Our
backbone, based on a reference Flan-T5-11B architecture, learns a universal
representation of the video that is a non-linear sum of the encoder models. It
learns using an autoregressive causal language modeling loss by predicting the
words spoken in YouTube videos. Finally, we evaluate on standard downstream
benchmarks by training fully connected prediction heads for each task. To the
best of our knowledge, this is the first use of multiple frozen SOTA models as
encoders in an "embedding -> backbone -> prediction head" design pattern - all
others have trained their own joint encoder models. Additionally, we include
more modalities than the current SOTA, Merlot Reserve, by adding explicit Scene
Graph information. For these two reasons, we believe it could combine the
world's best open-source models to achieve SOTA performance. Initial
experiments demonstrate the model is learning appropriately, but more
experimentation and compute is necessary, and already in progress, to realize
our loftier goals. Alongside this work, we build on the YT-20M dataset,
reproducing it and adding 25,000 personally selected YouTube videos to its
corpus. All code and model checkpoints are open sourced under a standard MIT
license.
|
2304.10508 | Perla Doubinsky | Perla Doubinsky (CEDRIC - VERTIGO, CNAM), Nicolas Audebert (CEDRIC -
VERTIGO, CNAM), Michel Crucianu (CEDRIC - VERTIGO), Herv\'e Le Borgne (CEA) | Wasserstein Loss for Semantic Editing in the Latent Space of GANs | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The latent space of GANs contains rich semantics reflecting the training
data. Different methods propose to learn edits in latent space corresponding to
semantic attributes, thus allowing to modify generated images. Most supervised
methods rely on the guidance of classifiers to produce such edits. However,
classifiers can lead to out-of-distribution regions and be fooled by
adversarial samples. We propose an alternative formulation based on the
Wasserstein loss that avoids such problems, while maintaining performance
on-par with classifier-based approaches. We demonstrate the effectiveness of
our method on two datasets (digits and faces) using StyleGAN2.
| [
{
"version": "v1",
"created": "Wed, 22 Mar 2023 08:15:27 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Doubinsky",
"Perla",
"",
"CEDRIC - VERTIGO, CNAM"
],
[
"Audebert",
"Nicolas",
"",
"CEDRIC -\n VERTIGO, CNAM"
],
[
"Crucianu",
"Michel",
"",
"CEDRIC - VERTIGO"
],
[
"Borgne",
"Hervé Le",
"",
"CEA"
]
] | TITLE: Wasserstein Loss for Semantic Editing in the Latent Space of GANs
ABSTRACT: The latent space of GANs contains rich semantics reflecting the training
data. Different methods propose to learn edits in latent space corresponding to
semantic attributes, thus allowing to modify generated images. Most supervised
methods rely on the guidance of classifiers to produce such edits. However,
classifiers can lead to out-of-distribution regions and be fooled by
adversarial samples. We propose an alternative formulation based on the
Wasserstein loss that avoids such problems, while maintaining performance
on-par with classifier-based approaches. We demonstrate the effectiveness of
our method on two datasets (digits and faces) using StyleGAN2.
|
2304.10510 | Andrew White | Quintina L. Campbell, Jonathan Herington, Andrew D. White | Censoring chemical data to mitigate dual use risk | null | null | null | null | cs.LG cs.CR cs.CY physics.chem-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The dual use of machine learning applications, where models can be used for
both beneficial and malicious purposes, presents a significant challenge. This
has recently become a particular concern in chemistry, where chemical datasets
containing sensitive labels (e.g. toxicological information) could be used to
develop predictive models that identify novel toxins or chemical warfare
agents. To mitigate dual use risks, we propose a model-agnostic method of
selectively noising datasets while preserving the utility of the data for
training deep neural networks in a beneficial region. We evaluate the
effectiveness of the proposed method across least squares, a multilayer
perceptron, and a graph neural network. Our findings show selectively noised
datasets can induce model variance and bias in predictions for sensitive labels
with control, suggesting the safe sharing of datasets containing sensitive
information is feasible. We also find omitting sensitive data often increases
model variance sufficiently to mitigate dual use. This work is proposed as a
foundation for future research on enabling more secure and collaborative data
sharing practices and safer machine learning applications in chemistry.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 17:46:30 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Campbell",
"Quintina L.",
""
],
[
"Herington",
"Jonathan",
""
],
[
"White",
"Andrew D.",
""
]
] | TITLE: Censoring chemical data to mitigate dual use risk
ABSTRACT: The dual use of machine learning applications, where models can be used for
both beneficial and malicious purposes, presents a significant challenge. This
has recently become a particular concern in chemistry, where chemical datasets
containing sensitive labels (e.g. toxicological information) could be used to
develop predictive models that identify novel toxins or chemical warfare
agents. To mitigate dual use risks, we propose a model-agnostic method of
selectively noising datasets while preserving the utility of the data for
training deep neural networks in a beneficial region. We evaluate the
effectiveness of the proposed method across least squares, a multilayer
perceptron, and a graph neural network. Our findings show selectively noised
datasets can induce model variance and bias in predictions for sensitive labels
with control, suggesting the safe sharing of datasets containing sensitive
information is feasible. We also find omitting sensitive data often increases
model variance sufficiently to mitigate dual use. This work is proposed as a
foundation for future research on enabling more secure and collaborative data
sharing practices and safer machine learning applications in chemistry.
|
2304.10511 | Hadi Eskandari | Hadi Eskandari, Michael Bewong, Sabih ur Rehman | OutCenTR: A novel semi-supervised framework for predicting exploits of
vulnerabilities in high-dimensional datasets | 11 pages, 6 figures | null | null | null | cs.CR cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | An ever-growing number of vulnerabilities are reported every day. Yet these
vulnerabilities are not all the same; Some are more targeted than others.
Correctly estimating the likelihood of a vulnerability being exploited is a
critical task for system administrators. This aids the system administrators in
prioritizing and patching the right vulnerabilities. Our work makes use of
outlier detection techniques to predict vulnerabilities that are likely to be
exploited in highly imbalanced and high-dimensional datasets such as the
National Vulnerability Database. We propose a dimensionality reduction
technique, OutCenTR, that enhances the baseline outlier detection models. We
further demonstrate the effectiveness and efficiency of OutCenTR empirically
with 4 benchmark and 12 synthetic datasets. The results of our experiments show
on average a 5-fold improvement of F1 score in comparison with state-of-the-art
dimensionality reduction techniques such as PCA and GRP.
| [
{
"version": "v1",
"created": "Mon, 3 Apr 2023 00:34:41 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Eskandari",
"Hadi",
""
],
[
"Bewong",
"Michael",
""
],
[
"Rehman",
"Sabih ur",
""
]
] | TITLE: OutCenTR: A novel semi-supervised framework for predicting exploits of
vulnerabilities in high-dimensional datasets
ABSTRACT: An ever-growing number of vulnerabilities are reported every day. Yet these
vulnerabilities are not all the same; Some are more targeted than others.
Correctly estimating the likelihood of a vulnerability being exploited is a
critical task for system administrators. This aids the system administrators in
prioritizing and patching the right vulnerabilities. Our work makes use of
outlier detection techniques to predict vulnerabilities that are likely to be
exploited in highly imbalanced and high-dimensional datasets such as the
National Vulnerability Database. We propose a dimensionality reduction
technique, OutCenTR, that enhances the baseline outlier detection models. We
further demonstrate the effectiveness and efficiency of OutCenTR empirically
with 4 benchmark and 12 synthetic datasets. The results of our experiments show
on average a 5-fold improvement of F1 score in comparison with state-of-the-art
dimensionality reduction techniques such as PCA and GRP.
|
2304.10515 | Lin Zhao | Lin Zhao, Haixing Dai, Zihao Wu, Dajiang Zhu, Tianming Liu | CP-CNN: Core-Periphery Principle Guided Convolutional Neural Network | null | null | null | null | cs.NE cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The evolution of convolutional neural networks (CNNs) can be largely
attributed to the design of its architecture, i.e., the network wiring pattern.
Neural architecture search (NAS) advances this by automating the search for the
optimal network architecture, but the resulting network instance may not
generalize well in different tasks. To overcome this, exploring network design
principles that are generalizable across tasks is a more practical solution. In
this study, We explore a novel brain-inspired design principle based on the
core-periphery property of the human brain network to guide the design of CNNs.
Our work draws inspiration from recent studies suggesting that artificial and
biological neural networks may have common principles in optimizing network
architecture. We implement the core-periphery principle in the design of
network wiring patterns and the sparsification of the convolution operation.
The resulting core-periphery principle guided CNNs (CP-CNNs) are evaluated on
three different datasets. The experiments demonstrate the effectiveness and
superiority compared to CNNs and ViT-based methods. Overall, our work
contributes to the growing field of brain-inspired AI by incorporating insights
from the human brain into the design of neural networks.
| [
{
"version": "v1",
"created": "Mon, 27 Mar 2023 03:59:43 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Zhao",
"Lin",
""
],
[
"Dai",
"Haixing",
""
],
[
"Wu",
"Zihao",
""
],
[
"Zhu",
"Dajiang",
""
],
[
"Liu",
"Tianming",
""
]
] | TITLE: CP-CNN: Core-Periphery Principle Guided Convolutional Neural Network
ABSTRACT: The evolution of convolutional neural networks (CNNs) can be largely
attributed to the design of its architecture, i.e., the network wiring pattern.
Neural architecture search (NAS) advances this by automating the search for the
optimal network architecture, but the resulting network instance may not
generalize well in different tasks. To overcome this, exploring network design
principles that are generalizable across tasks is a more practical solution. In
this study, We explore a novel brain-inspired design principle based on the
core-periphery property of the human brain network to guide the design of CNNs.
Our work draws inspiration from recent studies suggesting that artificial and
biological neural networks may have common principles in optimizing network
architecture. We implement the core-periphery principle in the design of
network wiring patterns and the sparsification of the convolution operation.
The resulting core-periphery principle guided CNNs (CP-CNNs) are evaluated on
three different datasets. The experiments demonstrate the effectiveness and
superiority compared to CNNs and ViT-based methods. Overall, our work
contributes to the growing field of brain-inspired AI by incorporating insights
from the human brain into the design of neural networks.
|
2304.10537 | Ziyu Wan | Ziyu Wan, Christian Richardt, Alja\v{z} Bo\v{z}i\v{c}, Chao Li, Vijay
Rengarajan, Seonghyeon Nam, Xiaoyu Xiang, Tuotuo Li, Bo Zhu, Rakesh Ranjan,
Jing Liao | Learning Neural Duplex Radiance Fields for Real-Time View Synthesis | CVPR 2023. Project page: http://raywzy.com/NDRF | null | null | null | cs.CV cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Neural radiance fields (NeRFs) enable novel view synthesis with unprecedented
visual quality. However, to render photorealistic images, NeRFs require
hundreds of deep multilayer perceptron (MLP) evaluations - for each pixel. This
is prohibitively expensive and makes real-time rendering infeasible, even on
powerful modern GPUs. In this paper, we propose a novel approach to distill and
bake NeRFs into highly efficient mesh-based neural representations that are
fully compatible with the massively parallel graphics rendering pipeline. We
represent scenes as neural radiance features encoded on a two-layer duplex
mesh, which effectively overcomes the inherent inaccuracies in 3D surface
reconstruction by learning the aggregated radiance information from a reliable
interval of ray-surface intersections. To exploit local geometric relationships
of nearby pixels, we leverage screen-space convolutions instead of the MLPs
used in NeRFs to achieve high-quality appearance. Finally, the performance of
the whole framework is further boosted by a novel multi-view distillation
optimization strategy. We demonstrate the effectiveness and superiority of our
approach via extensive experiments on a range of standard datasets.
| [
{
"version": "v1",
"created": "Thu, 20 Apr 2023 17:59:52 GMT"
}
] | 2023-04-21T00:00:00 | [
[
"Wan",
"Ziyu",
""
],
[
"Richardt",
"Christian",
""
],
[
"Božič",
"Aljaž",
""
],
[
"Li",
"Chao",
""
],
[
"Rengarajan",
"Vijay",
""
],
[
"Nam",
"Seonghyeon",
""
],
[
"Xiang",
"Xiaoyu",
""
],
[
"Li",
"Tuotuo",
""
],
[
"Zhu",
"Bo",
""
],
[
"Ranjan",
"Rakesh",
""
],
[
"Liao",
"Jing",
""
]
] | TITLE: Learning Neural Duplex Radiance Fields for Real-Time View Synthesis
ABSTRACT: Neural radiance fields (NeRFs) enable novel view synthesis with unprecedented
visual quality. However, to render photorealistic images, NeRFs require
hundreds of deep multilayer perceptron (MLP) evaluations - for each pixel. This
is prohibitively expensive and makes real-time rendering infeasible, even on
powerful modern GPUs. In this paper, we propose a novel approach to distill and
bake NeRFs into highly efficient mesh-based neural representations that are
fully compatible with the massively parallel graphics rendering pipeline. We
represent scenes as neural radiance features encoded on a two-layer duplex
mesh, which effectively overcomes the inherent inaccuracies in 3D surface
reconstruction by learning the aggregated radiance information from a reliable
interval of ray-surface intersections. To exploit local geometric relationships
of nearby pixels, we leverage screen-space convolutions instead of the MLPs
used in NeRFs to achieve high-quality appearance. Finally, the performance of
the whole framework is further boosted by a novel multi-view distillation
optimization strategy. We demonstrate the effectiveness and superiority of our
approach via extensive experiments on a range of standard datasets.
|
2106.07217 | Seulki Park | Seulki Park, Hwanjun Song, Daeho Um, Dae Ung Jo, Sangdoo Yun, and Jin
Young Choi | Influential Rank: A New Perspective of Post-training for Robust Model
against Noisy Labels | 15 pages | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep neural network can easily overfit to even noisy labels due to its high
capacity, which degrades the generalization performance of a model. To overcome
this issue, we propose a new approach for learning from noisy labels (LNL) via
post-training, which can significantly improve the generalization performance
of any pre-trained model on noisy label data. To this end, we rather exploit
the overfitting property of a trained model to identify mislabeled samples.
Specifically, our post-training approach gradually removes samples with high
influence on the decision boundary and refines the decision boundary to improve
generalization performance. Our post-training approach creates great synergies
when combined with the existing LNL methods. Experimental results on various
real-world and synthetic benchmark datasets demonstrate the validity of our
approach in diverse realistic scenarios.
| [
{
"version": "v1",
"created": "Mon, 14 Jun 2021 08:04:18 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Mar 2023 07:56:13 GMT"
},
{
"version": "v3",
"created": "Tue, 4 Apr 2023 12:36:53 GMT"
},
{
"version": "v4",
"created": "Wed, 19 Apr 2023 05:33:58 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Park",
"Seulki",
""
],
[
"Song",
"Hwanjun",
""
],
[
"Um",
"Daeho",
""
],
[
"Jo",
"Dae Ung",
""
],
[
"Yun",
"Sangdoo",
""
],
[
"Choi",
"Jin Young",
""
]
] | TITLE: Influential Rank: A New Perspective of Post-training for Robust Model
against Noisy Labels
ABSTRACT: Deep neural network can easily overfit to even noisy labels due to its high
capacity, which degrades the generalization performance of a model. To overcome
this issue, we propose a new approach for learning from noisy labels (LNL) via
post-training, which can significantly improve the generalization performance
of any pre-trained model on noisy label data. To this end, we rather exploit
the overfitting property of a trained model to identify mislabeled samples.
Specifically, our post-training approach gradually removes samples with high
influence on the decision boundary and refines the decision boundary to improve
generalization performance. Our post-training approach creates great synergies
when combined with the existing LNL methods. Experimental results on various
real-world and synthetic benchmark datasets demonstrate the validity of our
approach in diverse realistic scenarios.
|
2107.09287 | Hu Wang | Zihan Wang, Olivia Byrnes, Hu Wang, Ruoxi Sun, Congbo Ma, Huaming
Chen, Qi Wu, Minhui Xue | Data Hiding with Deep Learning: A Survey Unifying Digital Watermarking
and Steganography | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The advancement of secure communication and identity verification fields has
significantly increased through the use of deep learning techniques for data
hiding. By embedding information into a noise-tolerant signal such as audio,
video, or images, digital watermarking and steganography techniques can be used
to protect sensitive intellectual property and enable confidential
communication, ensuring that the information embedded is only accessible to
authorized parties. This survey provides an overview of recent developments in
deep learning techniques deployed for data hiding, categorized systematically
according to model architectures and noise injection methods. The objective
functions, evaluation metrics, and datasets used for training these data hiding
models are comprehensively summarised. Additionally, potential future research
directions that unite digital watermarking and steganography on software
engineering to enhance security and mitigate risks are suggested and
deliberated. This contribution furthers the creation of a more trustworthy
digital world and advances Responsible AI.
| [
{
"version": "v1",
"created": "Tue, 20 Jul 2021 07:03:23 GMT"
},
{
"version": "v2",
"created": "Sat, 25 Mar 2023 03:55:43 GMT"
},
{
"version": "v3",
"created": "Wed, 19 Apr 2023 05:09:08 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Wang",
"Zihan",
""
],
[
"Byrnes",
"Olivia",
""
],
[
"Wang",
"Hu",
""
],
[
"Sun",
"Ruoxi",
""
],
[
"Ma",
"Congbo",
""
],
[
"Chen",
"Huaming",
""
],
[
"Wu",
"Qi",
""
],
[
"Xue",
"Minhui",
""
]
] | TITLE: Data Hiding with Deep Learning: A Survey Unifying Digital Watermarking
and Steganography
ABSTRACT: The advancement of secure communication and identity verification fields has
significantly increased through the use of deep learning techniques for data
hiding. By embedding information into a noise-tolerant signal such as audio,
video, or images, digital watermarking and steganography techniques can be used
to protect sensitive intellectual property and enable confidential
communication, ensuring that the information embedded is only accessible to
authorized parties. This survey provides an overview of recent developments in
deep learning techniques deployed for data hiding, categorized systematically
according to model architectures and noise injection methods. The objective
functions, evaluation metrics, and datasets used for training these data hiding
models are comprehensively summarised. Additionally, potential future research
directions that unite digital watermarking and steganography on software
engineering to enhance security and mitigate risks are suggested and
deliberated. This contribution furthers the creation of a more trustworthy
digital world and advances Responsible AI.
|
2202.03936 | Madhurananda Pahar | Madhurananda Pahar, Igor Miranda, Andreas Diacon and Thomas Niesler | Accelerometer-based Bed Occupancy Detection for Automatic, Non-invasive
Long-term Cough Monitoring | null | IEEE Access, vol. 11, pp. 30739-30752, 2023 | 10.1109/ACCESS.2023.3261557 | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | We present a new machine learning based bed-occupancy detection system that
uses the accelerometer signal captured by a bed-attached consumer smartphone.
Automatic bed-occupancy detection is necessary for automatic long-term cough
monitoring, since the time which the monitored patient occupies the bed is
required to accurately calculate a cough rate. Accelerometer measurements are
more cost effective and less intrusive than alternatives such as video
monitoring or pressure sensors. A 249-hour dataset of manually-labelled
acceleration signals gathered from seven patients undergoing treatment for
tuberculosis (TB) was compiled for experimentation. These signals are
characterised by brief activity bursts interspersed with long periods of little
or no activity, even when the bed is occupied. To process them effectively, we
propose an architecture consisting of three interconnected components. An
occupancy-change detector locates instances at which bed occupancy is likely to
have changed, an occupancy-interval detector classifies periods between
detected occupancy changes and an occupancy-state detector corrects
falsely-identified occupancy changes. Using long short-term memory (LSTM)
networks, this architecture was demonstrated to achieve an AUC of 0.94. When
integrated into a complete cough monitoring system, the daily cough rate of a
patient undergoing TB treatment was determined over a period of 14 days. As the
colony forming unit (CFU) counts decreased and the time to positivity (TPP)
increased, the measured cough rate decreased, indicating effective TB
treatment. This provides a first indication that automatic cough monitoring
based on bed-mounted accelerometer measurements may present a non-invasive,
non-intrusive and cost-effective means of monitoring long-term recovery of TB
patients.
| [
{
"version": "v1",
"created": "Tue, 8 Feb 2022 15:38:34 GMT"
},
{
"version": "v2",
"created": "Sun, 13 Mar 2022 17:56:19 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Pahar",
"Madhurananda",
""
],
[
"Miranda",
"Igor",
""
],
[
"Diacon",
"Andreas",
""
],
[
"Niesler",
"Thomas",
""
]
] | TITLE: Accelerometer-based Bed Occupancy Detection for Automatic, Non-invasive
Long-term Cough Monitoring
ABSTRACT: We present a new machine learning based bed-occupancy detection system that
uses the accelerometer signal captured by a bed-attached consumer smartphone.
Automatic bed-occupancy detection is necessary for automatic long-term cough
monitoring, since the time which the monitored patient occupies the bed is
required to accurately calculate a cough rate. Accelerometer measurements are
more cost effective and less intrusive than alternatives such as video
monitoring or pressure sensors. A 249-hour dataset of manually-labelled
acceleration signals gathered from seven patients undergoing treatment for
tuberculosis (TB) was compiled for experimentation. These signals are
characterised by brief activity bursts interspersed with long periods of little
or no activity, even when the bed is occupied. To process them effectively, we
propose an architecture consisting of three interconnected components. An
occupancy-change detector locates instances at which bed occupancy is likely to
have changed, an occupancy-interval detector classifies periods between
detected occupancy changes and an occupancy-state detector corrects
falsely-identified occupancy changes. Using long short-term memory (LSTM)
networks, this architecture was demonstrated to achieve an AUC of 0.94. When
integrated into a complete cough monitoring system, the daily cough rate of a
patient undergoing TB treatment was determined over a period of 14 days. As the
colony forming unit (CFU) counts decreased and the time to positivity (TPP)
increased, the measured cough rate decreased, indicating effective TB
treatment. This provides a first indication that automatic cough monitoring
based on bed-mounted accelerometer measurements may present a non-invasive,
non-intrusive and cost-effective means of monitoring long-term recovery of TB
patients.
|
2202.08557 | Yinuo Zhao | Yinuo Zhao, Kun Wu, Zhiyuan Xu, Zhengping Che, Qi Lu, Jian Tang, Chi
Harold Liu | CADRE: A Cascade Deep Reinforcement Learning Framework for Vision-based
Autonomous Urban Driving | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Vision-based autonomous urban driving in dense traffic is quite challenging
due to the complicated urban environment and the dynamics of the driving
behaviors. Widely-applied methods either heavily rely on hand-crafted rules or
learn from limited human experience, which makes them hard to generalize to
rare but critical scenarios. In this paper, we present a novel CAscade Deep
REinforcement learning framework, CADRE, to achieve model-free vision-based
autonomous urban driving. In CADRE, to derive representative latent features
from raw observations, we first offline train a Co-attention Perception Module
(CoPM) that leverages the co-attention mechanism to learn the
inter-relationships between the visual and control information from a
pre-collected driving dataset. Cascaded by the frozen CoPM, we then present an
efficient distributed proximal policy optimization framework to online learn
the driving policy under the guidance of particularly designed reward
functions. We perform a comprehensive empirical study with the CARLA NoCrash
benchmark as well as specific obstacle avoidance scenarios in autonomous urban
driving tasks. The experimental results well justify the effectiveness of CADRE
and its superiority over the state-of-the-art by a wide margin.
| [
{
"version": "v1",
"created": "Thu, 17 Feb 2022 10:07:16 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 15:24:35 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Zhao",
"Yinuo",
""
],
[
"Wu",
"Kun",
""
],
[
"Xu",
"Zhiyuan",
""
],
[
"Che",
"Zhengping",
""
],
[
"Lu",
"Qi",
""
],
[
"Tang",
"Jian",
""
],
[
"Liu",
"Chi Harold",
""
]
] | TITLE: CADRE: A Cascade Deep Reinforcement Learning Framework for Vision-based
Autonomous Urban Driving
ABSTRACT: Vision-based autonomous urban driving in dense traffic is quite challenging
due to the complicated urban environment and the dynamics of the driving
behaviors. Widely-applied methods either heavily rely on hand-crafted rules or
learn from limited human experience, which makes them hard to generalize to
rare but critical scenarios. In this paper, we present a novel CAscade Deep
REinforcement learning framework, CADRE, to achieve model-free vision-based
autonomous urban driving. In CADRE, to derive representative latent features
from raw observations, we first offline train a Co-attention Perception Module
(CoPM) that leverages the co-attention mechanism to learn the
inter-relationships between the visual and control information from a
pre-collected driving dataset. Cascaded by the frozen CoPM, we then present an
efficient distributed proximal policy optimization framework to online learn
the driving policy under the guidance of particularly designed reward
functions. We perform a comprehensive empirical study with the CARLA NoCrash
benchmark as well as specific obstacle avoidance scenarios in autonomous urban
driving tasks. The experimental results well justify the effectiveness of CADRE
and its superiority over the state-of-the-art by a wide margin.
|
2203.01387 | Rafael Figueiredo Prudencio | Rafael Figueiredo Prudencio, Marcos R. O. A. Maximo, Esther Luna
Colombini | A Survey on Offline Reinforcement Learning: Taxonomy, Review, and Open
Problems | 21 pages; Final version accepted to IEEE Transactions on Neural
Networks and Learning Systems | null | 10.1109/TNNLS.2023.3250269 | null | cs.LG cs.AI stat.ML | http://creativecommons.org/licenses/by/4.0/ | With the widespread adoption of deep learning, reinforcement learning (RL)
has experienced a dramatic increase in popularity, scaling to previously
intractable problems, such as playing complex games from pixel observations,
sustaining conversations with humans, and controlling robotic agents. However,
there is still a wide range of domains inaccessible to RL due to the high cost
and danger of interacting with the environment. Offline RL is a paradigm that
learns exclusively from static datasets of previously collected interactions,
making it feasible to extract policies from large and diverse training
datasets. Effective offline RL algorithms have a much wider range of
applications than online RL, being particularly appealing for real-world
applications, such as education, healthcare, and robotics. In this work, we
contribute with a unifying taxonomy to classify offline RL methods.
Furthermore, we provide a comprehensive review of the latest algorithmic
breakthroughs in the field using a unified notation as well as a review of
existing benchmarks' properties and shortcomings. Additionally, we provide a
figure that summarizes the performance of each method and class of methods on
different dataset properties, equipping researchers with the tools to decide
which type of algorithm is best suited for the problem at hand and identify
which classes of algorithms look the most promising. Finally, we provide our
perspective on open problems and propose future research directions for this
rapidly growing field.
| [
{
"version": "v1",
"created": "Wed, 2 Mar 2022 20:05:11 GMT"
},
{
"version": "v2",
"created": "Sat, 5 Mar 2022 10:36:40 GMT"
},
{
"version": "v3",
"created": "Wed, 19 Apr 2023 00:30:20 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Prudencio",
"Rafael Figueiredo",
""
],
[
"Maximo",
"Marcos R. O. A.",
""
],
[
"Colombini",
"Esther Luna",
""
]
] | TITLE: A Survey on Offline Reinforcement Learning: Taxonomy, Review, and Open
Problems
ABSTRACT: With the widespread adoption of deep learning, reinforcement learning (RL)
has experienced a dramatic increase in popularity, scaling to previously
intractable problems, such as playing complex games from pixel observations,
sustaining conversations with humans, and controlling robotic agents. However,
there is still a wide range of domains inaccessible to RL due to the high cost
and danger of interacting with the environment. Offline RL is a paradigm that
learns exclusively from static datasets of previously collected interactions,
making it feasible to extract policies from large and diverse training
datasets. Effective offline RL algorithms have a much wider range of
applications than online RL, being particularly appealing for real-world
applications, such as education, healthcare, and robotics. In this work, we
contribute with a unifying taxonomy to classify offline RL methods.
Furthermore, we provide a comprehensive review of the latest algorithmic
breakthroughs in the field using a unified notation as well as a review of
existing benchmarks' properties and shortcomings. Additionally, we provide a
figure that summarizes the performance of each method and class of methods on
different dataset properties, equipping researchers with the tools to decide
which type of algorithm is best suited for the problem at hand and identify
which classes of algorithms look the most promising. Finally, we provide our
perspective on open problems and propose future research directions for this
rapidly growing field.
|
2204.02802 | Bla\v{z} Bertalani\v{c} | Bla\v{z} Bertalani\v{c}, Jakob Jenko and Carolina Fortuna | Dimensionality Expansion of Load Monitoring Time Series and Transfer
Learning for EMS | This paper has been withdrawn because it was significantly altered
and does not fit as a replacement anymore | null | null | null | cs.LG eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Energy management systems (EMS) rely on (non)-intrusive load monitoring
(N)ILM to monitor and manage appliances and help residents be more energy
efficient and thus more frugal. The robustness as well as the transfer
potential of the most promising machine learning solutions for (N)ILM is not
yet fully understood as they are trained and evaluated on relatively limited
data. In this paper, we propose a new approach for load monitoring in building
EMS based on dimensionality expansion of time series and transfer learning. We
perform an extensive evaluation on 5 different low-frequency datasets. The
proposed feature dimensionality expansion using video-like transformation and
resource-aware deep learning architecture achieves an average weighted F1 score
of 0.88 across the datasets with 29 appliances and is computationally more
efficient compared to the state-of-the-art imaging methods. Investigating the
proposed method for cross-dataset intra-domain transfer learning, we find that
1) our method performs with an average weighted F1 score of 0.80 while
requiring 3-times fewer epochs for model training compared to the non-transfer
approach, 2) can achieve an F1 score of 0.75 with only 230 data samples, and 3)
our transfer approach outperforms the state-of-the-art in precision drop by up
to 12 percentage points for unseen appliances.
| [
{
"version": "v1",
"created": "Wed, 6 Apr 2022 13:13:24 GMT"
},
{
"version": "v2",
"created": "Sat, 3 Sep 2022 10:52:11 GMT"
},
{
"version": "v3",
"created": "Wed, 25 Jan 2023 13:54:11 GMT"
},
{
"version": "v4",
"created": "Wed, 19 Apr 2023 10:09:37 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Bertalanič",
"Blaž",
""
],
[
"Jenko",
"Jakob",
""
],
[
"Fortuna",
"Carolina",
""
]
] | TITLE: Dimensionality Expansion of Load Monitoring Time Series and Transfer
Learning for EMS
ABSTRACT: Energy management systems (EMS) rely on (non)-intrusive load monitoring
(N)ILM to monitor and manage appliances and help residents be more energy
efficient and thus more frugal. The robustness as well as the transfer
potential of the most promising machine learning solutions for (N)ILM is not
yet fully understood as they are trained and evaluated on relatively limited
data. In this paper, we propose a new approach for load monitoring in building
EMS based on dimensionality expansion of time series and transfer learning. We
perform an extensive evaluation on 5 different low-frequency datasets. The
proposed feature dimensionality expansion using video-like transformation and
resource-aware deep learning architecture achieves an average weighted F1 score
of 0.88 across the datasets with 29 appliances and is computationally more
efficient compared to the state-of-the-art imaging methods. Investigating the
proposed method for cross-dataset intra-domain transfer learning, we find that
1) our method performs with an average weighted F1 score of 0.80 while
requiring 3-times fewer epochs for model training compared to the non-transfer
approach, 2) can achieve an F1 score of 0.75 with only 230 data samples, and 3)
our transfer approach outperforms the state-of-the-art in precision drop by up
to 12 percentage points for unseen appliances.
|
2205.13213 | Bohan Zhuang | Zizheng Pan, Jianfei Cai, Bohan Zhuang | Fast Vision Transformers with HiLo Attention | NeurIPS 2022 camera ready | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Vision Transformers (ViTs) have triggered the most recent and significant
breakthroughs in computer vision. Their efficient designs are mostly guided by
the indirect metric of computational complexity, i.e., FLOPs, which however has
a clear gap with the direct metric such as throughput. Thus, we propose to use
the direct speed evaluation on the target platform as the design principle for
efficient ViTs. Particularly, we introduce LITv2, a simple and effective ViT
which performs favourably against the existing state-of-the-art methods across
a spectrum of different model sizes with faster speed. At the core of LITv2 is
a novel self-attention mechanism, which we dub HiLo. HiLo is inspired by the
insight that high frequencies in an image capture local fine details and low
frequencies focus on global structures, whereas a multi-head self-attention
layer neglects the characteristic of different frequencies. Therefore, we
propose to disentangle the high/low frequency patterns in an attention layer by
separating the heads into two groups, where one group encodes high frequencies
via self-attention within each local window, and another group encodes low
frequencies by performing global attention between the average-pooled
low-frequency keys and values from each window and each query position in the
input feature map. Benefiting from the efficient design for both groups, we
show that HiLo is superior to the existing attention mechanisms by
comprehensively benchmarking FLOPs, speed and memory consumption on GPUs and
CPUs. For example, HiLo is 1.4x faster than spatial reduction attention and
1.6x faster than local window attention on CPUs. Powered by HiLo, LITv2 serves
as a strong backbone for mainstream vision tasks including image
classification, dense detection and segmentation. Code is available at
https://github.com/ziplab/LITv2.
| [
{
"version": "v1",
"created": "Thu, 26 May 2022 08:16:14 GMT"
},
{
"version": "v2",
"created": "Sat, 17 Sep 2022 06:24:35 GMT"
},
{
"version": "v3",
"created": "Sat, 15 Oct 2022 03:37:47 GMT"
},
{
"version": "v4",
"created": "Thu, 19 Jan 2023 01:12:54 GMT"
},
{
"version": "v5",
"created": "Wed, 19 Apr 2023 12:04:13 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Pan",
"Zizheng",
""
],
[
"Cai",
"Jianfei",
""
],
[
"Zhuang",
"Bohan",
""
]
] | TITLE: Fast Vision Transformers with HiLo Attention
ABSTRACT: Vision Transformers (ViTs) have triggered the most recent and significant
breakthroughs in computer vision. Their efficient designs are mostly guided by
the indirect metric of computational complexity, i.e., FLOPs, which however has
a clear gap with the direct metric such as throughput. Thus, we propose to use
the direct speed evaluation on the target platform as the design principle for
efficient ViTs. Particularly, we introduce LITv2, a simple and effective ViT
which performs favourably against the existing state-of-the-art methods across
a spectrum of different model sizes with faster speed. At the core of LITv2 is
a novel self-attention mechanism, which we dub HiLo. HiLo is inspired by the
insight that high frequencies in an image capture local fine details and low
frequencies focus on global structures, whereas a multi-head self-attention
layer neglects the characteristic of different frequencies. Therefore, we
propose to disentangle the high/low frequency patterns in an attention layer by
separating the heads into two groups, where one group encodes high frequencies
via self-attention within each local window, and another group encodes low
frequencies by performing global attention between the average-pooled
low-frequency keys and values from each window and each query position in the
input feature map. Benefiting from the efficient design for both groups, we
show that HiLo is superior to the existing attention mechanisms by
comprehensively benchmarking FLOPs, speed and memory consumption on GPUs and
CPUs. For example, HiLo is 1.4x faster than spatial reduction attention and
1.6x faster than local window attention on CPUs. Powered by HiLo, LITv2 serves
as a strong backbone for mainstream vision tasks including image
classification, dense detection and segmentation. Code is available at
https://github.com/ziplab/LITv2.
|
2206.04011 | Qiyu Hu | Qiyu Hu, Guangyi Zhang, Zhijin Qin, Yunlong Cai, Guanding Yu, Geoffrey
Ye Li | Robust Semantic Communications with Masked VQ-VAE Enabled Codebook | 16 pages, 11 figures. arXiv admin note: text overlap with
arXiv:2202.03338 | null | null | null | eess.SP cs.IT cs.LG math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although semantic communications have exhibited satisfactory performance for
a large number of tasks, the impact of semantic noise and the robustness of the
systems have not been well investigated. Semantic noise refers to the
misleading between the intended semantic symbols and received ones, thus cause
the failure of tasks. In this paper, we first propose a framework for the
robust end-to-end semantic communication systems to combat the semantic noise.
In particular, we analyze sample-dependent and sample-independent semantic
noise. To combat the semantic noise, the adversarial training with weight
perturbation is developed to incorporate the samples with semantic noise in the
training dataset. Then, we propose to mask a portion of the input, where the
semantic noise appears frequently, and design the masked vector
quantized-variational autoencoder (VQ-VAE) with the noise-related masking
strategy. We use a discrete codebook shared by the transmitter and the receiver
for encoded feature representation. To further improve the system robustness,
we develop a feature importance module (FIM) to suppress the noise-related and
task-unrelated features. Thus, the transmitter simply needs to transmit the
indices of these important task-related features in the codebook. Simulation
results show that the proposed method can be applied in many downstream tasks
and significantly improve the robustness against semantic noise with remarkable
reduction on the transmission overhead.
| [
{
"version": "v1",
"created": "Wed, 8 Jun 2022 16:58:47 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 02:55:11 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Hu",
"Qiyu",
""
],
[
"Zhang",
"Guangyi",
""
],
[
"Qin",
"Zhijin",
""
],
[
"Cai",
"Yunlong",
""
],
[
"Yu",
"Guanding",
""
],
[
"Li",
"Geoffrey Ye",
""
]
] | TITLE: Robust Semantic Communications with Masked VQ-VAE Enabled Codebook
ABSTRACT: Although semantic communications have exhibited satisfactory performance for
a large number of tasks, the impact of semantic noise and the robustness of the
systems have not been well investigated. Semantic noise refers to the
misleading between the intended semantic symbols and received ones, thus cause
the failure of tasks. In this paper, we first propose a framework for the
robust end-to-end semantic communication systems to combat the semantic noise.
In particular, we analyze sample-dependent and sample-independent semantic
noise. To combat the semantic noise, the adversarial training with weight
perturbation is developed to incorporate the samples with semantic noise in the
training dataset. Then, we propose to mask a portion of the input, where the
semantic noise appears frequently, and design the masked vector
quantized-variational autoencoder (VQ-VAE) with the noise-related masking
strategy. We use a discrete codebook shared by the transmitter and the receiver
for encoded feature representation. To further improve the system robustness,
we develop a feature importance module (FIM) to suppress the noise-related and
task-unrelated features. Thus, the transmitter simply needs to transmit the
indices of these important task-related features in the codebook. Simulation
results show that the proposed method can be applied in many downstream tasks
and significantly improve the robustness against semantic noise with remarkable
reduction on the transmission overhead.
|
2206.14409 | Zengqiang Yan | Xian Lin, Li Yu, Kwang-Ting Cheng, and Zengqiang Yan | BATFormer: Towards Boundary-Aware Lightweight Transformer for Efficient
Medical Image Segmentation | Accepted by IEEE Journal of Biomedical and Health Informatics The
source code is publicly available at https://github.com/xianlin7/BATFormer | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Objective: Transformers, born to remedy the inadequate receptive fields of
CNNs, have drawn explosive attention recently. However, the daunting
computational complexity of global representation learning, together with rigid
window partitioning, hinders their deployment in medical image segmentation.
This work aims to address the above two issues in transformers for better
medical image segmentation. Methods: We propose a boundary-aware lightweight
transformer (BATFormer) that can build cross-scale global interaction with
lower computational complexity and generate windows flexibly under the guidance
of entropy. Specifically, to fully explore the benefits of transformers in
long-range dependency establishment, a cross-scale global transformer (CGT)
module is introduced to jointly utilize multiple small-scale feature maps for
richer global features with lower computational complexity. Given the
importance of shape modeling in medical image segmentation, a boundary-aware
local transformer (BLT) module is constructed. Different from rigid window
partitioning in vanilla transformers which would produce boundary distortion,
BLT adopts an adaptive window partitioning scheme under the guidance of entropy
for both computational complexity reduction and shape preservation. Results:
BATFormer achieves the best performance in Dice of 92.84%, 91.97%, 90.26%, and
96.30% for the average, right ventricle, myocardium, and left ventricle
respectively on the ACDC dataset and the best performance in Dice, IoU, and ACC
of 90.76%, 84.64%, and 96.76% respectively on the ISIC 2018 dataset. More
importantly, BATFormer requires the least amount of model parameters and the
lowest computational complexity compared to the state-of-the-art approaches.
Conclusion and Significance: Our results demonstrate the necessity of
developing customized transformers for efficient and better medical image
segmentation.
| [
{
"version": "v1",
"created": "Wed, 29 Jun 2022 05:37:16 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Jul 2022 12:23:41 GMT"
},
{
"version": "v3",
"created": "Wed, 19 Apr 2023 02:43:34 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Lin",
"Xian",
""
],
[
"Yu",
"Li",
""
],
[
"Cheng",
"Kwang-Ting",
""
],
[
"Yan",
"Zengqiang",
""
]
] | TITLE: BATFormer: Towards Boundary-Aware Lightweight Transformer for Efficient
Medical Image Segmentation
ABSTRACT: Objective: Transformers, born to remedy the inadequate receptive fields of
CNNs, have drawn explosive attention recently. However, the daunting
computational complexity of global representation learning, together with rigid
window partitioning, hinders their deployment in medical image segmentation.
This work aims to address the above two issues in transformers for better
medical image segmentation. Methods: We propose a boundary-aware lightweight
transformer (BATFormer) that can build cross-scale global interaction with
lower computational complexity and generate windows flexibly under the guidance
of entropy. Specifically, to fully explore the benefits of transformers in
long-range dependency establishment, a cross-scale global transformer (CGT)
module is introduced to jointly utilize multiple small-scale feature maps for
richer global features with lower computational complexity. Given the
importance of shape modeling in medical image segmentation, a boundary-aware
local transformer (BLT) module is constructed. Different from rigid window
partitioning in vanilla transformers which would produce boundary distortion,
BLT adopts an adaptive window partitioning scheme under the guidance of entropy
for both computational complexity reduction and shape preservation. Results:
BATFormer achieves the best performance in Dice of 92.84%, 91.97%, 90.26%, and
96.30% for the average, right ventricle, myocardium, and left ventricle
respectively on the ACDC dataset and the best performance in Dice, IoU, and ACC
of 90.76%, 84.64%, and 96.76% respectively on the ISIC 2018 dataset. More
importantly, BATFormer requires the least amount of model parameters and the
lowest computational complexity compared to the state-of-the-art approaches.
Conclusion and Significance: Our results demonstrate the necessity of
developing customized transformers for efficient and better medical image
segmentation.
|
2207.01450 | Yinya Huang | Yinya Huang, Lemao Liu, Kun Xu, Meng Fang, Liang Lin, and Xiaodan
Liang | Discourse-Aware Graph Networks for Textual Logical Reasoning | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Textual logical reasoning, especially question-answering (QA) tasks with
logical reasoning, requires awareness of particular logical structures. The
passage-level logical relations represent entailment or contradiction between
propositional units (e.g., a concluding sentence). However, such structures are
unexplored as current QA systems focus on entity-based relations. In this work,
we propose logic structural-constraint modeling to solve the logical reasoning
QA and introduce discourse-aware graph networks (DAGNs). The networks first
construct logic graphs leveraging in-line discourse connectives and generic
logic theories, then learn logic representations by end-to-end evolving the
logic relations with an edge-reasoning mechanism and updating the graph
features. This pipeline is applied to a general encoder, whose fundamental
features are joined with the high-level logic features for answer prediction.
Experiments on three textual logical reasoning datasets demonstrate the
reasonability of the logical structures built in DAGNs and the effectiveness of
the learned logic features. Moreover, zero-shot transfer results show the
features' generality to unseen logical texts.
| [
{
"version": "v1",
"created": "Mon, 4 Jul 2022 14:38:49 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 03:53:59 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Huang",
"Yinya",
""
],
[
"Liu",
"Lemao",
""
],
[
"Xu",
"Kun",
""
],
[
"Fang",
"Meng",
""
],
[
"Lin",
"Liang",
""
],
[
"Liang",
"Xiaodan",
""
]
] | TITLE: Discourse-Aware Graph Networks for Textual Logical Reasoning
ABSTRACT: Textual logical reasoning, especially question-answering (QA) tasks with
logical reasoning, requires awareness of particular logical structures. The
passage-level logical relations represent entailment or contradiction between
propositional units (e.g., a concluding sentence). However, such structures are
unexplored as current QA systems focus on entity-based relations. In this work,
we propose logic structural-constraint modeling to solve the logical reasoning
QA and introduce discourse-aware graph networks (DAGNs). The networks first
construct logic graphs leveraging in-line discourse connectives and generic
logic theories, then learn logic representations by end-to-end evolving the
logic relations with an edge-reasoning mechanism and updating the graph
features. This pipeline is applied to a general encoder, whose fundamental
features are joined with the high-level logic features for answer prediction.
Experiments on three textual logical reasoning datasets demonstrate the
reasonability of the logical structures built in DAGNs and the effectiveness of
the learned logic features. Moreover, zero-shot transfer results show the
features' generality to unseen logical texts.
|
2208.05972 | Eshwar Jagadeesh Savitha | Eshwar J. Savitha and Roger A. Sauer | A new anisotropic bending model for nonlinear shells: Comparison with
existing models and isogeometric finite element implementation | null | null | 10.1016/j.ijsolstr.2023.112169 | null | cs.CE | http://creativecommons.org/licenses/by-nc-nd/4.0/ | A new nonlinear hyperelastic bending model for shells formulated directly in
surface form is presented, and compared to four prominently used bending
models. Through an essential set of elementary nonlinear bending test cases,
the stresses and moments of each model are examined analytically. Only the
proposed bending model passes all the test cases while the other bending models
either fail or only pass the test cases for small deformations. The proposed
new bending model can handle large deformations and initially curved surfaces.
It is based on the principal curvatures and their directions in the initial
configuration, and it thus can have different bending moduli along those
directions. These characteristics make it flexible in modeling a given
material, while it does not suffer from the pathologies of existing bending
models. Further, the bending models are compared computationally through four
classical benchmark examples and one contact example. As the underlying shell
theory is based on Kirchhoff-Love kinematics, isogeometric NURBS shape
functions are used to discretize the shell surface. The linearization and
efficient finite element implementation of the proposed new model are also
provided.
| [
{
"version": "v1",
"created": "Thu, 11 Aug 2022 17:33:41 GMT"
},
{
"version": "v2",
"created": "Tue, 4 Apr 2023 17:15:34 GMT"
},
{
"version": "v3",
"created": "Tue, 18 Apr 2023 19:54:13 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Savitha",
"Eshwar J.",
""
],
[
"Sauer",
"Roger A.",
""
]
] | TITLE: A new anisotropic bending model for nonlinear shells: Comparison with
existing models and isogeometric finite element implementation
ABSTRACT: A new nonlinear hyperelastic bending model for shells formulated directly in
surface form is presented, and compared to four prominently used bending
models. Through an essential set of elementary nonlinear bending test cases,
the stresses and moments of each model are examined analytically. Only the
proposed bending model passes all the test cases while the other bending models
either fail or only pass the test cases for small deformations. The proposed
new bending model can handle large deformations and initially curved surfaces.
It is based on the principal curvatures and their directions in the initial
configuration, and it thus can have different bending moduli along those
directions. These characteristics make it flexible in modeling a given
material, while it does not suffer from the pathologies of existing bending
models. Further, the bending models are compared computationally through four
classical benchmark examples and one contact example. As the underlying shell
theory is based on Kirchhoff-Love kinematics, isogeometric NURBS shape
functions are used to discretize the shell surface. The linearization and
efficient finite element implementation of the proposed new model are also
provided.
|
2209.03715 | Soner Candas | Soner Candas, Beneharo Reveron Baecker, Anurag Mohapatra, Thomas
Hamacher | Optimization-based framework for low-voltage grid reinforcement
assessment under various levels of flexibility and coordination | null | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | The rapid electrification of residential heating and mobility sectors is
expected to drive the existing distribution grid assets beyond their planned
operating conditions. This change will also reveal new potentials through
sector coupling, flexibilities, and the local exchange of decentralized
generation. This paper thus presents an optimization framework for multi-modal
energy systems at the low voltage (LV) distribution grid level. In this, we
focus on the reinforcement requirements of the grid and the techno-economic
assessment of flexibility components and coordination between agents. By
employing a multi-level approach, computational complexity is reduced, and
various levels of coordination and flexibilities are implemented. We conclude
the work with a case study for a representative rural grid in Germany, in which
we observe high economic potential in the flexible operation of buildings,
majorly thanks to better integration of photovoltaics. Across all paradigms
barring a best-case benchmark, grid reinforcement based on a mix of passive and
active measures was necessary. A synergy effect is observed between
flexibilities and coordination, as their combination reduces the peaks to the
extent of completely avoiding grid reinforcement. The presented framework can
be applied with a wide range of grid and component types to outline the broad
landscape of future LV distribution grids.
| [
{
"version": "v1",
"created": "Thu, 8 Sep 2022 11:19:09 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 15:06:26 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Candas",
"Soner",
""
],
[
"Baecker",
"Beneharo Reveron",
""
],
[
"Mohapatra",
"Anurag",
""
],
[
"Hamacher",
"Thomas",
""
]
] | TITLE: Optimization-based framework for low-voltage grid reinforcement
assessment under various levels of flexibility and coordination
ABSTRACT: The rapid electrification of residential heating and mobility sectors is
expected to drive the existing distribution grid assets beyond their planned
operating conditions. This change will also reveal new potentials through
sector coupling, flexibilities, and the local exchange of decentralized
generation. This paper thus presents an optimization framework for multi-modal
energy systems at the low voltage (LV) distribution grid level. In this, we
focus on the reinforcement requirements of the grid and the techno-economic
assessment of flexibility components and coordination between agents. By
employing a multi-level approach, computational complexity is reduced, and
various levels of coordination and flexibilities are implemented. We conclude
the work with a case study for a representative rural grid in Germany, in which
we observe high economic potential in the flexible operation of buildings,
majorly thanks to better integration of photovoltaics. Across all paradigms
barring a best-case benchmark, grid reinforcement based on a mix of passive and
active measures was necessary. A synergy effect is observed between
flexibilities and coordination, as their combination reduces the peaks to the
extent of completely avoiding grid reinforcement. The presented framework can
be applied with a wide range of grid and component types to outline the broad
landscape of future LV distribution grids.
|
2209.14569 | Chenxin An | Chenxin An, Ming Zhong, Zhiyong Wu, Qin Zhu, Xuanjing Huang, Xipeng
Qiu | COLO: A Contrastive Learning based Re-ranking Framework for One-Stage
Summarization | Accepted by COLING 2022 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Traditional training paradigms for extractive and abstractive summarization
systems always only use token-level or sentence-level training objectives.
However, the output summary is always evaluated from summary-level which leads
to the inconsistency in training and evaluation. In this paper, we propose a
Contrastive Learning based re-ranking framework for one-stage summarization
called COLO. By modeling a contrastive objective, we show that the
summarization model is able to directly generate summaries according to the
summary-level score without additional modules and parameters. Extensive
experiments demonstrate that COLO boosts the extractive and abstractive results
of one-stage systems on CNN/DailyMail benchmark to 44.58 and 46.33 ROUGE-1
score while preserving the parameter efficiency and inference efficiency.
Compared with state-of-the-art multi-stage systems, we save more than 100 GPU
training hours and obtaining 3~8 speed-up ratio during inference while
maintaining comparable results.
| [
{
"version": "v1",
"created": "Thu, 29 Sep 2022 06:11:21 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 07:01:18 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"An",
"Chenxin",
""
],
[
"Zhong",
"Ming",
""
],
[
"Wu",
"Zhiyong",
""
],
[
"Zhu",
"Qin",
""
],
[
"Huang",
"Xuanjing",
""
],
[
"Qiu",
"Xipeng",
""
]
] | TITLE: COLO: A Contrastive Learning based Re-ranking Framework for One-Stage
Summarization
ABSTRACT: Traditional training paradigms for extractive and abstractive summarization
systems always only use token-level or sentence-level training objectives.
However, the output summary is always evaluated from summary-level which leads
to the inconsistency in training and evaluation. In this paper, we propose a
Contrastive Learning based re-ranking framework for one-stage summarization
called COLO. By modeling a contrastive objective, we show that the
summarization model is able to directly generate summaries according to the
summary-level score without additional modules and parameters. Extensive
experiments demonstrate that COLO boosts the extractive and abstractive results
of one-stage systems on CNN/DailyMail benchmark to 44.58 and 46.33 ROUGE-1
score while preserving the parameter efficiency and inference efficiency.
Compared with state-of-the-art multi-stage systems, we save more than 100 GPU
training hours and obtaining 3~8 speed-up ratio during inference while
maintaining comparable results.
|
2210.02897 | Jithin Jagannath | Anu Jagannath and Jithin Jagannath | Embedding-Assisted Attentional Deep Learning for Real-World RF
Fingerprinting of Bluetooth | To Appear in IEEE Transactions on Cognitive Communications and
Networking | null | null | null | cs.NI cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | A scalable and computationally efficient framework is designed to fingerprint
real-world Bluetooth devices. We propose an embedding-assisted attentional
framework (Mbed-ATN) suitable for fingerprinting actual Bluetooth devices. Its
generalization capability is analyzed in different settings and the effect of
sample length and anti-aliasing decimation is demonstrated. The embedding
module serves as a dimensionality reduction unit that maps the high dimensional
3D input tensor to a 1D feature vector for further processing by the ATN
module. Furthermore, unlike the prior research in this field, we closely
evaluate the complexity of the model and test its fingerprinting capability
with real-world Bluetooth dataset collected under a different time frame and
experimental setting while being trained on another. Our study reveals a 9.17x
and 65.2x lesser memory usage at a sample length of 100 kS when compared to the
benchmark - GRU and Oracle models respectively. Further, the proposed Mbed-ATN
showcases 16.9x fewer FLOPs and 7.5x lesser trainable parameters when compared
to Oracle. Finally, we show that when subject to anti-aliasing decimation and
at greater input sample lengths of 1 MS, the proposed Mbed-ATN framework
results in a 5.32x higher TPR, 37.9% fewer false alarms, and 6.74x higher
accuracy under the challenging real-world setting.
| [
{
"version": "v1",
"created": "Thu, 22 Sep 2022 17:06:13 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 12:57:50 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Jagannath",
"Anu",
""
],
[
"Jagannath",
"Jithin",
""
]
] | TITLE: Embedding-Assisted Attentional Deep Learning for Real-World RF
Fingerprinting of Bluetooth
ABSTRACT: A scalable and computationally efficient framework is designed to fingerprint
real-world Bluetooth devices. We propose an embedding-assisted attentional
framework (Mbed-ATN) suitable for fingerprinting actual Bluetooth devices. Its
generalization capability is analyzed in different settings and the effect of
sample length and anti-aliasing decimation is demonstrated. The embedding
module serves as a dimensionality reduction unit that maps the high dimensional
3D input tensor to a 1D feature vector for further processing by the ATN
module. Furthermore, unlike the prior research in this field, we closely
evaluate the complexity of the model and test its fingerprinting capability
with real-world Bluetooth dataset collected under a different time frame and
experimental setting while being trained on another. Our study reveals a 9.17x
and 65.2x lesser memory usage at a sample length of 100 kS when compared to the
benchmark - GRU and Oracle models respectively. Further, the proposed Mbed-ATN
showcases 16.9x fewer FLOPs and 7.5x lesser trainable parameters when compared
to Oracle. Finally, we show that when subject to anti-aliasing decimation and
at greater input sample lengths of 1 MS, the proposed Mbed-ATN framework
results in a 5.32x higher TPR, 37.9% fewer false alarms, and 6.74x higher
accuracy under the challenging real-world setting.
|
2210.05038 | Pedro Rodriguez | Pedro Rodriguez, Mahmoud Azab, Becka Silvert, Renato Sanchez, Linzy
Labson, Hardik Shah and Seungwhan Moon | Fighting FIRe with FIRE: Assessing the Validity of Text-to-Video
Retrieval Benchmarks | EACL 2023 Camera Ready | null | null | null | cs.CL cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Searching troves of videos with textual descriptions is a core multimodal
retrieval task. Owing to the lack of a purpose-built dataset for text-to-video
retrieval, video captioning datasets have been re-purposed to evaluate models
by (1) treating captions as positive matches to their respective videos and (2)
assuming all other videos to be negatives. However, this methodology leads to a
fundamental flaw during evaluation: since captions are marked as relevant only
to their original video, many alternate videos also match the caption, which
introduces false-negative caption-video pairs. We show that when these false
negatives are corrected, a recent state-of-the-art model gains 25\% recall
points -- a difference that threatens the validity of the benchmark itself. To
diagnose and mitigate this issue, we annotate and release 683K additional
caption-video pairs. Using these, we recompute effectiveness scores for three
models on two standard benchmarks (MSR-VTT and MSVD). We find that (1) the
recomputed metrics are up to 25\% recall points higher for the best models, (2)
these benchmarks are nearing saturation for Recall@10, (3) caption length
(generality) is related to the number of positives, and (4) annotation costs
can be mitigated through sampling. We recommend retiring these benchmarks in
their current form, and we make recommendations for future text-to-video
retrieval benchmarks.
| [
{
"version": "v1",
"created": "Mon, 10 Oct 2022 22:45:06 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 03:50:48 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Rodriguez",
"Pedro",
""
],
[
"Azab",
"Mahmoud",
""
],
[
"Silvert",
"Becka",
""
],
[
"Sanchez",
"Renato",
""
],
[
"Labson",
"Linzy",
""
],
[
"Shah",
"Hardik",
""
],
[
"Moon",
"Seungwhan",
""
]
] | TITLE: Fighting FIRe with FIRE: Assessing the Validity of Text-to-Video
Retrieval Benchmarks
ABSTRACT: Searching troves of videos with textual descriptions is a core multimodal
retrieval task. Owing to the lack of a purpose-built dataset for text-to-video
retrieval, video captioning datasets have been re-purposed to evaluate models
by (1) treating captions as positive matches to their respective videos and (2)
assuming all other videos to be negatives. However, this methodology leads to a
fundamental flaw during evaluation: since captions are marked as relevant only
to their original video, many alternate videos also match the caption, which
introduces false-negative caption-video pairs. We show that when these false
negatives are corrected, a recent state-of-the-art model gains 25\% recall
points -- a difference that threatens the validity of the benchmark itself. To
diagnose and mitigate this issue, we annotate and release 683K additional
caption-video pairs. Using these, we recompute effectiveness scores for three
models on two standard benchmarks (MSR-VTT and MSVD). We find that (1) the
recomputed metrics are up to 25\% recall points higher for the best models, (2)
these benchmarks are nearing saturation for Recall@10, (3) caption length
(generality) is related to the number of positives, and (4) annotation costs
can be mitigated through sampling. We recommend retiring these benchmarks in
their current form, and we make recommendations for future text-to-video
retrieval benchmarks.
|
2210.06386 | Lang Feng | Lang Feng, Qianhui Liu, Huajin Tang, De Ma, Gang Pan | Multi-Level Firing with Spiking DS-ResNet: Enabling Better and Deeper
Directly-Trained Spiking Neural Networks | Accepted by the Thirty-First International Joint Conference on
Artificial Intelligence (IJCAI-22) | null | null | null | cs.NE cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spiking neural networks (SNNs) are bio-inspired neural networks with
asynchronous discrete and sparse characteristics, which have increasingly
manifested their superiority in low energy consumption. Recent research is
devoted to utilizing spatio-temporal information to directly train SNNs by
backpropagation. However, the binary and non-differentiable properties of spike
activities force directly trained SNNs to suffer from serious gradient
vanishing and network degradation, which greatly limits the performance of
directly trained SNNs and prevents them from going deeper. In this paper, we
propose a multi-level firing (MLF) method based on the existing spatio-temporal
back propagation (STBP) method, and spiking dormant-suppressed residual network
(spiking DS-ResNet). MLF enables more efficient gradient propagation and the
incremental expression ability of the neurons. Spiking DS-ResNet can
efficiently perform identity mapping of discrete spikes, as well as provide a
more suitable connection for gradient propagation in deep SNNs. With the
proposed method, our model achieves superior performances on a non-neuromorphic
dataset and two neuromorphic datasets with much fewer trainable parameters and
demonstrates the great ability to combat the gradient vanishing and degradation
problem in deep SNNs.
| [
{
"version": "v1",
"created": "Wed, 12 Oct 2022 16:39:46 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 09:09:13 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Feng",
"Lang",
""
],
[
"Liu",
"Qianhui",
""
],
[
"Tang",
"Huajin",
""
],
[
"Ma",
"De",
""
],
[
"Pan",
"Gang",
""
]
] | TITLE: Multi-Level Firing with Spiking DS-ResNet: Enabling Better and Deeper
Directly-Trained Spiking Neural Networks
ABSTRACT: Spiking neural networks (SNNs) are bio-inspired neural networks with
asynchronous discrete and sparse characteristics, which have increasingly
manifested their superiority in low energy consumption. Recent research is
devoted to utilizing spatio-temporal information to directly train SNNs by
backpropagation. However, the binary and non-differentiable properties of spike
activities force directly trained SNNs to suffer from serious gradient
vanishing and network degradation, which greatly limits the performance of
directly trained SNNs and prevents them from going deeper. In this paper, we
propose a multi-level firing (MLF) method based on the existing spatio-temporal
back propagation (STBP) method, and spiking dormant-suppressed residual network
(spiking DS-ResNet). MLF enables more efficient gradient propagation and the
incremental expression ability of the neurons. Spiking DS-ResNet can
efficiently perform identity mapping of discrete spikes, as well as provide a
more suitable connection for gradient propagation in deep SNNs. With the
proposed method, our model achieves superior performances on a non-neuromorphic
dataset and two neuromorphic datasets with much fewer trainable parameters and
demonstrates the great ability to combat the gradient vanishing and degradation
problem in deep SNNs.
|
2210.07558 | Mojtaba Valipour | Mojtaba Valipour, Mehdi Rezagholizadeh, Ivan Kobyzev, Ali Ghodsi | DyLoRA: Parameter Efficient Tuning of Pre-trained Models using Dynamic
Search-Free Low-Rank Adaptation | Accepted to EACL 2023 | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | With the ever-growing size of pretrained models (PMs), fine-tuning them has
become more expensive and resource-hungry. As a remedy, low-rank adapters
(LoRA) keep the main pretrained weights of the model frozen and just introduce
some learnable truncated SVD modules (so-called LoRA blocks) to the model.
While LoRA blocks are parameter-efficient, they suffer from two major problems:
first, the size of these blocks is fixed and cannot be modified after training
(for example, if we need to change the rank of LoRA blocks, then we need to
re-train them from scratch); second, optimizing their rank requires an
exhaustive search and effort. In this work, we introduce a dynamic low-rank
adaptation (DyLoRA) technique to address these two problems together. Our
DyLoRA method trains LoRA blocks for a range of ranks instead of a single rank
by sorting the representation learned by the adapter module at different ranks
during training. We evaluate our solution on different natural language
understanding (GLUE benchmark) and language generation tasks (E2E, DART and
WebNLG) using different pretrained models such as RoBERTa and GPT with
different sizes. Our results show that we can train dynamic search-free models
with DyLoRA at least 4 to 7 times (depending to the task) faster than LoRA
without significantly compromising performance. Moreover, our models can
perform consistently well on a much larger range of ranks compared to LoRA.
| [
{
"version": "v1",
"created": "Fri, 14 Oct 2022 06:29:22 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 04:28:24 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Valipour",
"Mojtaba",
""
],
[
"Rezagholizadeh",
"Mehdi",
""
],
[
"Kobyzev",
"Ivan",
""
],
[
"Ghodsi",
"Ali",
""
]
] | TITLE: DyLoRA: Parameter Efficient Tuning of Pre-trained Models using Dynamic
Search-Free Low-Rank Adaptation
ABSTRACT: With the ever-growing size of pretrained models (PMs), fine-tuning them has
become more expensive and resource-hungry. As a remedy, low-rank adapters
(LoRA) keep the main pretrained weights of the model frozen and just introduce
some learnable truncated SVD modules (so-called LoRA blocks) to the model.
While LoRA blocks are parameter-efficient, they suffer from two major problems:
first, the size of these blocks is fixed and cannot be modified after training
(for example, if we need to change the rank of LoRA blocks, then we need to
re-train them from scratch); second, optimizing their rank requires an
exhaustive search and effort. In this work, we introduce a dynamic low-rank
adaptation (DyLoRA) technique to address these two problems together. Our
DyLoRA method trains LoRA blocks for a range of ranks instead of a single rank
by sorting the representation learned by the adapter module at different ranks
during training. We evaluate our solution on different natural language
understanding (GLUE benchmark) and language generation tasks (E2E, DART and
WebNLG) using different pretrained models such as RoBERTa and GPT with
different sizes. Our results show that we can train dynamic search-free models
with DyLoRA at least 4 to 7 times (depending to the task) faster than LoRA
without significantly compromising performance. Moreover, our models can
perform consistently well on a much larger range of ranks compared to LoRA.
|
2210.13605 | Samrudhdhi B Rangrej | Samrudhdhi B Rangrej, Kevin J Liang, Tal Hassner, James J Clark | GliTr: Glimpse Transformers with Spatiotemporal Consistency for Online
Action Prediction | Accepted to WACV 2023 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Many online action prediction models observe complete frames to locate and
attend to informative subregions in the frames called glimpses and recognize an
ongoing action based on global and local information. However, in applications
with constrained resources, an agent may not be able to observe the complete
frame, yet must still locate useful glimpses to predict an incomplete action
based on local information only. In this paper, we develop Glimpse Transformers
(GliTr), which observe only narrow glimpses at all times, thus predicting an
ongoing action and the following most informative glimpse location based on the
partial spatiotemporal information collected so far. In the absence of a ground
truth for the optimal glimpse locations for action recognition, we train GliTr
using a novel spatiotemporal consistency objective: We require GliTr to attend
to the glimpses with features similar to the corresponding complete frames
(i.e. spatial consistency) and the resultant class logits at time $t$
equivalent to the ones predicted using whole frames up to $t$ (i.e. temporal
consistency). Inclusion of our proposed consistency objective yields ~10%
higher accuracy on the Something-Something-v2 (SSv2) dataset than the baseline
cross-entropy objective. Overall, despite observing only ~33% of the total area
per frame, GliTr achieves 53.02% and 93.91% accuracy on the SSv2 and Jester
datasets, respectively.
| [
{
"version": "v1",
"created": "Mon, 24 Oct 2022 21:10:34 GMT"
},
{
"version": "v2",
"created": "Wed, 19 Apr 2023 00:41:39 GMT"
}
] | 2023-04-20T00:00:00 | [
[
"Rangrej",
"Samrudhdhi B",
""
],
[
"Liang",
"Kevin J",
""
],
[
"Hassner",
"Tal",
""
],
[
"Clark",
"James J",
""
]
] | TITLE: GliTr: Glimpse Transformers with Spatiotemporal Consistency for Online
Action Prediction
ABSTRACT: Many online action prediction models observe complete frames to locate and
attend to informative subregions in the frames called glimpses and recognize an
ongoing action based on global and local information. However, in applications
with constrained resources, an agent may not be able to observe the complete
frame, yet must still locate useful glimpses to predict an incomplete action
based on local information only. In this paper, we develop Glimpse Transformers
(GliTr), which observe only narrow glimpses at all times, thus predicting an
ongoing action and the following most informative glimpse location based on the
partial spatiotemporal information collected so far. In the absence of a ground
truth for the optimal glimpse locations for action recognition, we train GliTr
using a novel spatiotemporal consistency objective: We require GliTr to attend
to the glimpses with features similar to the corresponding complete frames
(i.e. spatial consistency) and the resultant class logits at time $t$
equivalent to the ones predicted using whole frames up to $t$ (i.e. temporal
consistency). Inclusion of our proposed consistency objective yields ~10%
higher accuracy on the Something-Something-v2 (SSv2) dataset than the baseline
cross-entropy objective. Overall, despite observing only ~33% of the total area
per frame, GliTr achieves 53.02% and 93.91% accuracy on the SSv2 and Jester
datasets, respectively.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.