id
large_stringlengths
9
16
submitter
large_stringlengths
3
64
authors
large_stringlengths
5
20.2k
title
large_stringlengths
5
245
comments
large_stringlengths
1
482
journal-ref
large_stringlengths
4
382
doi
large_stringlengths
9
151
report-no
large_stringlengths
2
509
categories
large_stringlengths
5
78
license
large_stringclasses
9 values
abstract
large_stringlengths
83
3.55k
versions
large listlengths
1
21
update_date
timestamp[ms]date
2007-05-23 00:00:00
2025-05-30 00:00:00
authors_parsed
large listlengths
1
1.35k
prompt
large_stringlengths
166
3.64k
label
stringclasses
2 values
prob
float64
0.5
0.73
2108.02566
Min Yang
Yufeng Wang, Dan Li, Cong Xu, Min Yang
Missingness Augmentation: A General Approach for Improving Generative Imputation Models
20 pages
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Missing data imputation is a fundamental problem in data analysis, and many studies have been conducted to improve its performance by exploring model structures and learning procedures. However, data augmentation, as a simple yet effective method, has not received enough attention in this area. In this paper, we propose a novel data augmentation method called Missingness Augmentation (MisA) for generative imputation models. Our approach dynamically produces incomplete samples at each epoch by utilizing the generator's output, constraining the augmented samples using a simple reconstruction loss, and combining this loss with the original loss to form the final optimization objective. As a general augmentation technique, MisA can be easily integrated into generative imputation frameworks, providing a simple yet effective way to enhance their performance. Experimental results demonstrate that MisA significantly improves the performance of many recently proposed generative imputation models on a variety of tabular and image datasets. The code is available at \url{https://github.com/WYu-Feng/Missingness-Augmentation}.
[ { "version": "v1", "created": "Sat, 31 Jul 2021 08:51:46 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 06:05:14 GMT" } ]
2023-04-07T00:00:00
[ [ "Wang", "Yufeng", "" ], [ "Li", "Dan", "" ], [ "Xu", "Cong", "" ], [ "Yang", "Min", "" ] ]
TITLE: Missingness Augmentation: A General Approach for Improving Generative Imputation Models ABSTRACT: Missing data imputation is a fundamental problem in data analysis, and many studies have been conducted to improve its performance by exploring model structures and learning procedures. However, data augmentation, as a simple yet effective method, has not received enough attention in this area. In this paper, we propose a novel data augmentation method called Missingness Augmentation (MisA) for generative imputation models. Our approach dynamically produces incomplete samples at each epoch by utilizing the generator's output, constraining the augmented samples using a simple reconstruction loss, and combining this loss with the original loss to form the final optimization objective. As a general augmentation technique, MisA can be easily integrated into generative imputation frameworks, providing a simple yet effective way to enhance their performance. Experimental results demonstrate that MisA significantly improves the performance of many recently proposed generative imputation models on a variety of tabular and image datasets. The code is available at \url{https://github.com/WYu-Feng/Missingness-Augmentation}.
no_new_dataset
0.709636
2110.11084
Haokui Zhang
Xizhe Xue, Haokui Zhang, Bei Fang, Zongwen Bai, Ying Li
Grafting Transformer on Automatically Designed Convolutional Neural Network for Hyperspectral Image Classification
15 pages, 10 figures
null
10.1109/TGRS.2022.3180685
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Hyperspectral image (HSI) classification has been a hot topic for decides, as hyperspectral images have rich spatial and spectral information and provide strong basis for distinguishing different land-cover objects. Benefiting from the development of deep learning technologies, deep learning based HSI classification methods have achieved promising performance. Recently, several neural architecture search (NAS) algorithms have been proposed for HSI classification, which further improve the accuracy of HSI classification to a new level. In this paper, NAS and Transformer are combined for handling HSI classification task for the first time. Compared with previous work, the proposed method has two main differences. First, we revisit the search spaces designed in previous HSI classification NAS methods and propose a novel hybrid search space, consisting of the space dominated cell and the spectrum dominated cell. Compared with search spaces proposed in previous works, the proposed hybrid search space is more aligned with the characteristic of HSI data, that is, HSIs have a relatively low spatial resolution and an extremely high spectral resolution. Second, to further improve the classification accuracy, we attempt to graft the emerging transformer module on the automatically designed convolutional neural network (CNN) to add global information to local region focused features learned by CNN. Experimental results on three public HSI datasets show that the proposed method achieves much better performance than comparison approaches, including manually designed network and NAS based HSI classification methods. Especially on the most recently captured dataset Houston University, overall accuracy is improved by nearly 6 percentage points. Code is available at: https://github.com/Cecilia-xue/HyT-NAS.
[ { "version": "v1", "created": "Thu, 21 Oct 2021 11:51:51 GMT" }, { "version": "v2", "created": "Fri, 10 Jun 2022 12:44:51 GMT" }, { "version": "v3", "created": "Thu, 6 Apr 2023 12:30:49 GMT" } ]
2023-04-07T00:00:00
[ [ "Xue", "Xizhe", "" ], [ "Zhang", "Haokui", "" ], [ "Fang", "Bei", "" ], [ "Bai", "Zongwen", "" ], [ "Li", "Ying", "" ] ]
TITLE: Grafting Transformer on Automatically Designed Convolutional Neural Network for Hyperspectral Image Classification ABSTRACT: Hyperspectral image (HSI) classification has been a hot topic for decides, as hyperspectral images have rich spatial and spectral information and provide strong basis for distinguishing different land-cover objects. Benefiting from the development of deep learning technologies, deep learning based HSI classification methods have achieved promising performance. Recently, several neural architecture search (NAS) algorithms have been proposed for HSI classification, which further improve the accuracy of HSI classification to a new level. In this paper, NAS and Transformer are combined for handling HSI classification task for the first time. Compared with previous work, the proposed method has two main differences. First, we revisit the search spaces designed in previous HSI classification NAS methods and propose a novel hybrid search space, consisting of the space dominated cell and the spectrum dominated cell. Compared with search spaces proposed in previous works, the proposed hybrid search space is more aligned with the characteristic of HSI data, that is, HSIs have a relatively low spatial resolution and an extremely high spectral resolution. Second, to further improve the classification accuracy, we attempt to graft the emerging transformer module on the automatically designed convolutional neural network (CNN) to add global information to local region focused features learned by CNN. Experimental results on three public HSI datasets show that the proposed method achieves much better performance than comparison approaches, including manually designed network and NAS based HSI classification methods. Especially on the most recently captured dataset Houston University, overall accuracy is improved by nearly 6 percentage points. Code is available at: https://github.com/Cecilia-xue/HyT-NAS.
no_new_dataset
0.713045
2201.11494
Kohei Watabe
Kohei Watabe, Shohei Nakazawa, Yoshiki Sato, Sho Tsugawa, Kenji Nakagawa
GraphTune: A Learning-based Graph Generative Model with Tunable Structural Features
The paper was published in IEEE Transactions on Network Science and Engineering (2023). An earlier and short version of this paper was presented at the 41st IEEE International Conference on Distributed Computing Systems (ICDCS 2021) Poster Track
null
10.1109/TNSE.2023.3244590
null
cs.LG cs.NI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Generative models for graphs have been actively studied for decades, and they have a wide range of applications. Recently, learning-based graph generation that reproduces real-world graphs has been attracting the attention of many researchers. Although several generative models that utilize modern machine learning technologies have been proposed, conditional generation of general graphs has been less explored in the field. In this paper, we propose a generative model that allows us to tune the value of a global-level structural feature as a condition. Our model, called GraphTune, makes it possible to tune the value of any structural feature of generated graphs using Long Short Term Memory (LSTM) and a Conditional Variational AutoEncoder (CVAE). We performed comparative evaluations of GraphTune and conventional models on a real graph dataset. The evaluations show that GraphTune makes it possible to more clearly tune the value of a global-level structural feature better than conventional models.
[ { "version": "v1", "created": "Thu, 27 Jan 2022 13:14:53 GMT" }, { "version": "v2", "created": "Mon, 21 Feb 2022 10:20:47 GMT" }, { "version": "v3", "created": "Wed, 5 Apr 2023 10:39:46 GMT" } ]
2023-04-07T00:00:00
[ [ "Watabe", "Kohei", "" ], [ "Nakazawa", "Shohei", "" ], [ "Sato", "Yoshiki", "" ], [ "Tsugawa", "Sho", "" ], [ "Nakagawa", "Kenji", "" ] ]
TITLE: GraphTune: A Learning-based Graph Generative Model with Tunable Structural Features ABSTRACT: Generative models for graphs have been actively studied for decades, and they have a wide range of applications. Recently, learning-based graph generation that reproduces real-world graphs has been attracting the attention of many researchers. Although several generative models that utilize modern machine learning technologies have been proposed, conditional generation of general graphs has been less explored in the field. In this paper, we propose a generative model that allows us to tune the value of a global-level structural feature as a condition. Our model, called GraphTune, makes it possible to tune the value of any structural feature of generated graphs using Long Short Term Memory (LSTM) and a Conditional Variational AutoEncoder (CVAE). We performed comparative evaluations of GraphTune and conventional models on a real graph dataset. The evaluations show that GraphTune makes it possible to more clearly tune the value of a global-level structural feature better than conventional models.
no_new_dataset
0.709629
2202.00582
Shaymaa Abdulla Al-Delayel
Shaymaa Abdulla Al-Delayel
Security Analysis of Mobile Banking Application in Qatar
8 pages, 1 figure
null
null
null
cs.CR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper discusses the security posture of Android m-banking applications in Qatar. Since technology has developed over the years and more security methods are provided, banking is now heavily reliant on mobile applications for prompt service delivery to clients, thus enabling a seamless and remote transaction. However, such mobile banking applications have access to sensitive data for each bank customer which presents a potential attack vector for clients, and the banks. The banks, therefore, have the responsibility to protect the information of the client by providing a high-security layer to their mobile application. This research discusses m-banking applications for Android OS, its security, vulnerability, threats, and solutions. Two m-banking applications were analyzed and benchmarked against standardized best practices, using the combination of two mobile testing frameworks. The security weaknesses observed during the experimental evaluation suggest the need for a more robust security evaluation of a mobile banking application in the state of Qatar. Such an approach would further ensure the confidence of the end-users. Consequently, understanding the security posture would provide a veritable measure towards mbanking security and user awareness.
[ { "version": "v1", "created": "Mon, 31 Jan 2022 17:52:30 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 09:25:37 GMT" } ]
2023-04-07T00:00:00
[ [ "Al-Delayel", "Shaymaa Abdulla", "" ] ]
TITLE: Security Analysis of Mobile Banking Application in Qatar ABSTRACT: This paper discusses the security posture of Android m-banking applications in Qatar. Since technology has developed over the years and more security methods are provided, banking is now heavily reliant on mobile applications for prompt service delivery to clients, thus enabling a seamless and remote transaction. However, such mobile banking applications have access to sensitive data for each bank customer which presents a potential attack vector for clients, and the banks. The banks, therefore, have the responsibility to protect the information of the client by providing a high-security layer to their mobile application. This research discusses m-banking applications for Android OS, its security, vulnerability, threats, and solutions. Two m-banking applications were analyzed and benchmarked against standardized best practices, using the combination of two mobile testing frameworks. The security weaknesses observed during the experimental evaluation suggest the need for a more robust security evaluation of a mobile banking application in the state of Qatar. Such an approach would further ensure the confidence of the end-users. Consequently, understanding the security posture would provide a veritable measure towards mbanking security and user awareness.
no_new_dataset
0.709044
2202.01470
Chunhua Shen
Guangkai Xu, Wei Yin, Hao Chen, Chunhua Shen, Kai Cheng, Feng Wu, Feng Zhao
Towards 3D Scene Reconstruction from Locally Scale-Aligned Monocular Video Depth
Technical error
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Existing monocular depth estimation methods have achieved excellent robustness in diverse scenes, but they can only retrieve affine-invariant depth, up to an unknown scale and shift. However, in some video-based scenarios such as video depth estimation and 3D scene reconstruction from a video, the unknown scale and shift residing in per-frame prediction may cause the depth inconsistency. To solve this problem, we propose a locally weighted linear regression method to recover the scale and shift with very sparse anchor points, which ensures the scale consistency along consecutive frames. Extensive experiments show that our method can boost the performance of existing state-of-the-art approaches by 50% at most over several zero-shot benchmarks. Besides, we merge over 6.3 million RGBD images to train strong and robust depth models. Our produced ResNet50-backbone model even outperforms the state-of-the-art DPT ViT-Large model. Combining with geometry-based reconstruction methods, we formulate a new dense 3D scene reconstruction pipeline, which benefits from both the scale consistency of sparse points and the robustness of monocular methods. By performing the simple per-frame prediction over a video, the accurate 3D scene shape can be recovered.
[ { "version": "v1", "created": "Thu, 3 Feb 2022 08:52:54 GMT" }, { "version": "v2", "created": "Fri, 22 Apr 2022 12:30:42 GMT" }, { "version": "v3", "created": "Mon, 5 Sep 2022 02:20:58 GMT" }, { "version": "v4", "created": "Thu, 6 Apr 2023 03:08:03 GMT" } ]
2023-04-07T00:00:00
[ [ "Xu", "Guangkai", "" ], [ "Yin", "Wei", "" ], [ "Chen", "Hao", "" ], [ "Shen", "Chunhua", "" ], [ "Cheng", "Kai", "" ], [ "Wu", "Feng", "" ], [ "Zhao", "Feng", "" ] ]
TITLE: Towards 3D Scene Reconstruction from Locally Scale-Aligned Monocular Video Depth ABSTRACT: Existing monocular depth estimation methods have achieved excellent robustness in diverse scenes, but they can only retrieve affine-invariant depth, up to an unknown scale and shift. However, in some video-based scenarios such as video depth estimation and 3D scene reconstruction from a video, the unknown scale and shift residing in per-frame prediction may cause the depth inconsistency. To solve this problem, we propose a locally weighted linear regression method to recover the scale and shift with very sparse anchor points, which ensures the scale consistency along consecutive frames. Extensive experiments show that our method can boost the performance of existing state-of-the-art approaches by 50% at most over several zero-shot benchmarks. Besides, we merge over 6.3 million RGBD images to train strong and robust depth models. Our produced ResNet50-backbone model even outperforms the state-of-the-art DPT ViT-Large model. Combining with geometry-based reconstruction methods, we formulate a new dense 3D scene reconstruction pipeline, which benefits from both the scale consistency of sparse points and the robustness of monocular methods. By performing the simple per-frame prediction over a video, the accurate 3D scene shape can be recovered.
no_new_dataset
0.70724
2202.10400
Nika Mansouri Ghiasi
Nika Mansouri Ghiasi, Jisung Park, Harun Mustafa, Jeremie Kim, Ataberk Olgun, Arvid Gollwitzer, Damla Senol Cali, Can Firtina, Haiyu Mao, Nour Almadhoun Alserr, Rachata Ausavarungnirun, Nandita Vijaykumar, Mohammed Alser, and Onur Mutlu
GenStore: A High-Performance and Energy-Efficient In-Storage Computing System for Genome Sequence Analysis
Published at ASPLOS 2022
null
null
null
cs.AR cs.DC cs.OS q-bio.GN
http://creativecommons.org/licenses/by/4.0/
Read mapping is a fundamental, yet computationally-expensive step in many genomics applications. It is used to identify potential matches and differences between fragments (called reads) of a sequenced genome and an already known genome (called a reference genome). To address the computational challenges in genome analysis, many prior works propose various approaches such as filters that select the reads that must undergo expensive computation, efficient heuristics, and hardware acceleration. While effective at reducing the computation overhead, all such approaches still require the costly movement of a large amount of data from storage to the rest of the system, which can significantly lower the end-to-end performance of read mapping in conventional and emerging genomics systems. We propose GenStore, the first in-storage processing system designed for genome sequence analysis that greatly reduces both data movement and computational overheads of genome sequence analysis by exploiting low-cost and accurate in-storage filters. GenStore leverages hardware/software co-design to address the challenges of in-storage processing, supporting reads with 1) different read lengths and error rates, and 2) different degrees of genetic variation. Through rigorous analysis of read mapping processes, we meticulously design low-cost hardware accelerators and data/computation flows inside a NAND flash-based SSD. Our evaluation using a wide range of real genomic datasets shows that GenStore, when implemented in three modern SSDs, significantly improves the read mapping performance of state-of-the-art software (hardware) baselines by 2.07-6.05$\times$ (1.52-3.32$\times$) for read sets with high similarity to the reference genome and 1.45-33.63$\times$ (2.70-19.2$\times$) for read sets with low similarity to the reference genome.
[ { "version": "v1", "created": "Mon, 21 Feb 2022 17:53:01 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 16:56:04 GMT" } ]
2023-04-07T00:00:00
[ [ "Ghiasi", "Nika Mansouri", "" ], [ "Park", "Jisung", "" ], [ "Mustafa", "Harun", "" ], [ "Kim", "Jeremie", "" ], [ "Olgun", "Ataberk", "" ], [ "Gollwitzer", "Arvid", "" ], [ "Cali", "Damla Senol", "" ], [ "Firtina", "Can", "" ], [ "Mao", "Haiyu", "" ], [ "Alserr", "Nour Almadhoun", "" ], [ "Ausavarungnirun", "Rachata", "" ], [ "Vijaykumar", "Nandita", "" ], [ "Alser", "Mohammed", "" ], [ "Mutlu", "Onur", "" ] ]
TITLE: GenStore: A High-Performance and Energy-Efficient In-Storage Computing System for Genome Sequence Analysis ABSTRACT: Read mapping is a fundamental, yet computationally-expensive step in many genomics applications. It is used to identify potential matches and differences between fragments (called reads) of a sequenced genome and an already known genome (called a reference genome). To address the computational challenges in genome analysis, many prior works propose various approaches such as filters that select the reads that must undergo expensive computation, efficient heuristics, and hardware acceleration. While effective at reducing the computation overhead, all such approaches still require the costly movement of a large amount of data from storage to the rest of the system, which can significantly lower the end-to-end performance of read mapping in conventional and emerging genomics systems. We propose GenStore, the first in-storage processing system designed for genome sequence analysis that greatly reduces both data movement and computational overheads of genome sequence analysis by exploiting low-cost and accurate in-storage filters. GenStore leverages hardware/software co-design to address the challenges of in-storage processing, supporting reads with 1) different read lengths and error rates, and 2) different degrees of genetic variation. Through rigorous analysis of read mapping processes, we meticulously design low-cost hardware accelerators and data/computation flows inside a NAND flash-based SSD. Our evaluation using a wide range of real genomic datasets shows that GenStore, when implemented in three modern SSDs, significantly improves the read mapping performance of state-of-the-art software (hardware) baselines by 2.07-6.05$\times$ (1.52-3.32$\times$) for read sets with high similarity to the reference genome and 1.45-33.63$\times$ (2.70-19.2$\times$) for read sets with low similarity to the reference genome.
no_new_dataset
0.709982
2203.09978
Jean-Christophe Gagnon-Audet
Jean-Christophe Gagnon-Audet, Kartik Ahuja, Mohammad-Javad Darvishi-Bayazi, Pooneh Mousavi, Guillaume Dumas, Irina Rish
WOODS: Benchmarks for Out-of-Distribution Generalization in Time Series
47 pages, 21 figures
null
null
null
cs.LG stat.ML
http://creativecommons.org/licenses/by/4.0/
Machine learning models often fail to generalize well under distributional shifts. Understanding and overcoming these failures have led to a research field of Out-of-Distribution (OOD) generalization. Despite being extensively studied for static computer vision tasks, OOD generalization has been underexplored for time series tasks. To shine light on this gap, we present WOODS: eight challenging open-source time series benchmarks covering a diverse range of data modalities, such as videos, brain recordings, and sensor signals. We revise the existing OOD generalization algorithms for time series tasks and evaluate them using our systematic framework. Our experiments show a large room for improvement for empirical risk minimization and OOD generalization algorithms on our datasets, thus underscoring the new challenges posed by time series tasks. Code and documentation are available at https://woods-benchmarks.github.io .
[ { "version": "v1", "created": "Fri, 18 Mar 2022 14:12:54 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 14:21:18 GMT" } ]
2023-04-07T00:00:00
[ [ "Gagnon-Audet", "Jean-Christophe", "" ], [ "Ahuja", "Kartik", "" ], [ "Darvishi-Bayazi", "Mohammad-Javad", "" ], [ "Mousavi", "Pooneh", "" ], [ "Dumas", "Guillaume", "" ], [ "Rish", "Irina", "" ] ]
TITLE: WOODS: Benchmarks for Out-of-Distribution Generalization in Time Series ABSTRACT: Machine learning models often fail to generalize well under distributional shifts. Understanding and overcoming these failures have led to a research field of Out-of-Distribution (OOD) generalization. Despite being extensively studied for static computer vision tasks, OOD generalization has been underexplored for time series tasks. To shine light on this gap, we present WOODS: eight challenging open-source time series benchmarks covering a diverse range of data modalities, such as videos, brain recordings, and sensor signals. We revise the existing OOD generalization algorithms for time series tasks and evaluate them using our systematic framework. Our experiments show a large room for improvement for empirical risk minimization and OOD generalization algorithms on our datasets, thus underscoring the new challenges posed by time series tasks. Code and documentation are available at https://woods-benchmarks.github.io .
no_new_dataset
0.525125
2205.05345
Michael Baur
Michael Baur, Benedikt Fesl, Michael Koller, Wolfgang Utschick
Variational Autoencoder Leveraged MMSE Channel Estimation
Accepted for publication at Asilomar 2022
null
10.1109/IEEECONF56349.2022.10051858
null
eess.SP cs.IT math.IT stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We propose to utilize a variational autoencoder (VAE) for data-driven channel estimation. The underlying true and unknown channel distribution is modeled by the VAE as a conditional Gaussian distribution in a novel way, parameterized by the respective first and second order conditional moments. As a result, it can be observed that the linear minimum mean square error (LMMSE) estimator in its variant conditioned on the latent sample of the VAE approximates an optimal MSE estimator. Furthermore, we argue how a VAE-based channel estimator can approximate the MMSE channel estimator. We propose three variants of VAE estimators that differ in the data used during training and estimation. First, we show that given perfectly known channel state information at the input of the VAE during estimation, which is impractical, we obtain an estimator that can serve as a benchmark result for an estimation scenario. We then propose practically feasible approaches, where perfectly known channel state information is only necessary in the training phase or is not needed at all. Simulation results on 3GPP and QuaDRiGa channel data attest a small performance loss of the practical approaches and the superiority of our VAE approaches in comparison to other related channel estimation methods.
[ { "version": "v1", "created": "Wed, 11 May 2022 08:45:54 GMT" }, { "version": "v2", "created": "Tue, 22 Nov 2022 14:32:20 GMT" } ]
2023-04-07T00:00:00
[ [ "Baur", "Michael", "" ], [ "Fesl", "Benedikt", "" ], [ "Koller", "Michael", "" ], [ "Utschick", "Wolfgang", "" ] ]
TITLE: Variational Autoencoder Leveraged MMSE Channel Estimation ABSTRACT: We propose to utilize a variational autoencoder (VAE) for data-driven channel estimation. The underlying true and unknown channel distribution is modeled by the VAE as a conditional Gaussian distribution in a novel way, parameterized by the respective first and second order conditional moments. As a result, it can be observed that the linear minimum mean square error (LMMSE) estimator in its variant conditioned on the latent sample of the VAE approximates an optimal MSE estimator. Furthermore, we argue how a VAE-based channel estimator can approximate the MMSE channel estimator. We propose three variants of VAE estimators that differ in the data used during training and estimation. First, we show that given perfectly known channel state information at the input of the VAE during estimation, which is impractical, we obtain an estimator that can serve as a benchmark result for an estimation scenario. We then propose practically feasible approaches, where perfectly known channel state information is only necessary in the training phase or is not needed at all. Simulation results on 3GPP and QuaDRiGa channel data attest a small performance loss of the practical approaches and the superiority of our VAE approaches in comparison to other related channel estimation methods.
no_new_dataset
0.708616
2205.05398
Francesca Cairoli
Luca Bortolussi, Francesca Cairoli, Ginevra Carbone, Paolo Pulcini
Scalable Stochastic Parametric Verification with Stochastic Variational Smoothed Model Checking
null
null
null
null
cs.LG stat.ML
http://creativecommons.org/licenses/by/4.0/
Parametric verification of linear temporal properties for stochastic models can be expressed as computing the satisfaction probability of a certain property as a function of the parameters of the model. Smoothed model checking (smMC) aims at inferring the satisfaction function over the entire parameter space from a limited set of observations obtained via simulation. As observations are costly and noisy, smMC is framed as a Bayesian inference problem so that the estimates have an additional quantification of the uncertainty. In smMC the authors use Gaussian Processes (GP), inferred by means of the Expectation Propagation algorithm. This approach provides accurate reconstructions with statistically sound quantification of the uncertainty. However, it inherits the well-known scalability issues of GP. In this paper, we exploit recent advances in probabilistic machine learning to push this limitation forward, making Bayesian inference of smMC scalable to larger datasets and enabling its application to models with high dimensional parameter spaces. We propose Stochastic Variational Smoothed Model Checking (SV-smMC), a solution that exploits stochastic variational inference (SVI) to approximate the posterior distribution of the smMC problem. The strength and flexibility of SVI make SV-smMC applicable to two alternative probabilistic models: Gaussian Processes (GP) and Bayesian Neural Networks (BNN). The core ingredient of SVI is a stochastic gradient-based optimization that makes inference easily parallelizable and that enables GPU acceleration. In this paper, we compare the performances of smMC against those of SV-smMC by looking at the scalability, the computational efficiency and the accuracy of the reconstructed satisfaction function.
[ { "version": "v1", "created": "Wed, 11 May 2022 10:43:23 GMT" }, { "version": "v2", "created": "Fri, 14 Oct 2022 12:38:45 GMT" }, { "version": "v3", "created": "Thu, 6 Apr 2023 10:53:49 GMT" } ]
2023-04-07T00:00:00
[ [ "Bortolussi", "Luca", "" ], [ "Cairoli", "Francesca", "" ], [ "Carbone", "Ginevra", "" ], [ "Pulcini", "Paolo", "" ] ]
TITLE: Scalable Stochastic Parametric Verification with Stochastic Variational Smoothed Model Checking ABSTRACT: Parametric verification of linear temporal properties for stochastic models can be expressed as computing the satisfaction probability of a certain property as a function of the parameters of the model. Smoothed model checking (smMC) aims at inferring the satisfaction function over the entire parameter space from a limited set of observations obtained via simulation. As observations are costly and noisy, smMC is framed as a Bayesian inference problem so that the estimates have an additional quantification of the uncertainty. In smMC the authors use Gaussian Processes (GP), inferred by means of the Expectation Propagation algorithm. This approach provides accurate reconstructions with statistically sound quantification of the uncertainty. However, it inherits the well-known scalability issues of GP. In this paper, we exploit recent advances in probabilistic machine learning to push this limitation forward, making Bayesian inference of smMC scalable to larger datasets and enabling its application to models with high dimensional parameter spaces. We propose Stochastic Variational Smoothed Model Checking (SV-smMC), a solution that exploits stochastic variational inference (SVI) to approximate the posterior distribution of the smMC problem. The strength and flexibility of SVI make SV-smMC applicable to two alternative probabilistic models: Gaussian Processes (GP) and Bayesian Neural Networks (BNN). The core ingredient of SVI is a stochastic gradient-based optimization that makes inference easily parallelizable and that enables GPU acceleration. In this paper, we compare the performances of smMC against those of SV-smMC by looking at the scalability, the computational efficiency and the accuracy of the reconstructed satisfaction function.
no_new_dataset
0.707392
2205.08076
Jaeyoun You
Jaeyoun You, Daemin Park, Joo-yeong Song, Bongwon Suh
A Labeling Task Design for Supporting Algorithmic Needs: Facilitating Worker Diversity and Reducing AI Bias
45 pages, 4 figures
null
10.1109/BigData55660.2022.10020415
null
cs.HC
http://creativecommons.org/licenses/by/4.0/
Studies on supervised machine learning (ML) recommend involving workers from various backgrounds in training dataset labeling to reduce algorithmic bias. Moreover, sophisticated tasks for categorizing objects in images are necessary to improve ML performance, further complicating micro-tasks. This study aims to develop a task design incorporating the fair participation of people, regardless of their specific backgrounds or task's difficulty. By collaborating with 75 labelers from diverse backgrounds for 3 months, we analyzed workers' log-data and relevant narratives to identify the task's hurdles and helpers. The findings revealed that workers' decision-making tendencies varied depending on their backgrounds. We found that the community that positively helps workers and the machine's feedback perceived by workers could make people easily engaged in works. Hence, ML's bias could be expectedly mitigated. Based on these findings, we suggest an extended human-in-the-loop approach that connects labelers, machines, and communities rather than isolating individual workers.
[ { "version": "v1", "created": "Tue, 17 May 2022 03:41:09 GMT" } ]
2023-04-07T00:00:00
[ [ "You", "Jaeyoun", "" ], [ "Park", "Daemin", "" ], [ "Song", "Joo-yeong", "" ], [ "Suh", "Bongwon", "" ] ]
TITLE: A Labeling Task Design for Supporting Algorithmic Needs: Facilitating Worker Diversity and Reducing AI Bias ABSTRACT: Studies on supervised machine learning (ML) recommend involving workers from various backgrounds in training dataset labeling to reduce algorithmic bias. Moreover, sophisticated tasks for categorizing objects in images are necessary to improve ML performance, further complicating micro-tasks. This study aims to develop a task design incorporating the fair participation of people, regardless of their specific backgrounds or task's difficulty. By collaborating with 75 labelers from diverse backgrounds for 3 months, we analyzed workers' log-data and relevant narratives to identify the task's hurdles and helpers. The findings revealed that workers' decision-making tendencies varied depending on their backgrounds. We found that the community that positively helps workers and the machine's feedback perceived by workers could make people easily engaged in works. Hence, ML's bias could be expectedly mitigated. Based on these findings, we suggest an extended human-in-the-loop approach that connects labelers, machines, and communities rather than isolating individual workers.
no_new_dataset
0.71222
2206.04516
Jaemin Yoo
Jaemin Yoo, Hyunsik Jeon, Jinhong Jung, and U Kang
Accurate Node Feature Estimation with Structured Variational Graph Autoencoder
Fixed Table 4 from the KDD 2022 version
null
null
null
cs.LG
http://creativecommons.org/licenses/by-nc-sa/4.0/
Given a graph with partial observations of node features, how can we estimate the missing features accurately? Feature estimation is a crucial problem for analyzing real-world graphs whose features are commonly missing during the data collection process. Accurate estimation not only provides diverse information of nodes but also supports the inference of graph neural networks that require the full observation of node features. However, designing an effective approach for estimating high-dimensional features is challenging, since it requires an estimator to have large representation power, increasing the risk of overfitting. In this work, we propose SVGA (Structured Variational Graph Autoencoder), an accurate method for feature estimation. SVGA applies strong regularization to the distribution of latent variables by structured variational inference, which models the prior of variables as Gaussian Markov random field based on the graph structure. As a result, SVGA combines the advantages of probabilistic inference and graph neural networks, achieving state-of-the-art performance in real datasets.
[ { "version": "v1", "created": "Thu, 9 Jun 2022 14:07:45 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 00:55:18 GMT" } ]
2023-04-07T00:00:00
[ [ "Yoo", "Jaemin", "" ], [ "Jeon", "Hyunsik", "" ], [ "Jung", "Jinhong", "" ], [ "Kang", "U", "" ] ]
TITLE: Accurate Node Feature Estimation with Structured Variational Graph Autoencoder ABSTRACT: Given a graph with partial observations of node features, how can we estimate the missing features accurately? Feature estimation is a crucial problem for analyzing real-world graphs whose features are commonly missing during the data collection process. Accurate estimation not only provides diverse information of nodes but also supports the inference of graph neural networks that require the full observation of node features. However, designing an effective approach for estimating high-dimensional features is challenging, since it requires an estimator to have large representation power, increasing the risk of overfitting. In this work, we propose SVGA (Structured Variational Graph Autoencoder), an accurate method for feature estimation. SVGA applies strong regularization to the distribution of latent variables by structured variational inference, which models the prior of variables as Gaussian Markov random field based on the graph structure. As a result, SVGA combines the advantages of probabilistic inference and graph neural networks, achieving state-of-the-art performance in real datasets.
no_new_dataset
0.708862
2207.00422
An Yan
An Yan, Zhankui He, Jiacheng Li, Tianyang Zhang, Julian McAuley
Personalized Showcases: Generating Multi-Modal Explanations for Recommendations
Accepted to SIGIR-23, with additional dataset details. Code and data: https://github.com/zzxslp/Gest
null
null
null
cs.IR cs.AI cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Existing explanation models generate only text for recommendations but still struggle to produce diverse contents. In this paper, to further enrich explanations, we propose a new task named personalized showcases, in which we provide both textual and visual information to explain our recommendations. Specifically, we first select a personalized image set that is the most relevant to a user's interest toward a recommended item. Then, natural language explanations are generated accordingly given our selected images. For this new task, we collect a large-scale dataset from Google Local (i.e.,~maps) and construct a high-quality subset for generating multi-modal explanations. We propose a personalized multi-modal framework which can generate diverse and visually-aligned explanations via contrastive learning. Experiments show that our framework benefits from different modalities as inputs, and is able to produce more diverse and expressive explanations compared to previous methods on a variety of evaluation metrics.
[ { "version": "v1", "created": "Thu, 30 Jun 2022 01:43:58 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 17:59:43 GMT" } ]
2023-04-07T00:00:00
[ [ "Yan", "An", "" ], [ "He", "Zhankui", "" ], [ "Li", "Jiacheng", "" ], [ "Zhang", "Tianyang", "" ], [ "McAuley", "Julian", "" ] ]
TITLE: Personalized Showcases: Generating Multi-Modal Explanations for Recommendations ABSTRACT: Existing explanation models generate only text for recommendations but still struggle to produce diverse contents. In this paper, to further enrich explanations, we propose a new task named personalized showcases, in which we provide both textual and visual information to explain our recommendations. Specifically, we first select a personalized image set that is the most relevant to a user's interest toward a recommended item. Then, natural language explanations are generated accordingly given our selected images. For this new task, we collect a large-scale dataset from Google Local (i.e.,~maps) and construct a high-quality subset for generating multi-modal explanations. We propose a personalized multi-modal framework which can generate diverse and visually-aligned explanations via contrastive learning. Experiments show that our framework benefits from different modalities as inputs, and is able to produce more diverse and expressive explanations compared to previous methods on a variety of evaluation metrics.
no_new_dataset
0.70133
2207.07278
Mengyin Liu
Mengyin Liu, Chao Zhu, Hongyu Gao, Weibo Gu, Hongfa Wang, Wei Liu, Xu-cheng Yin
Boosting Multi-Modal E-commerce Attribute Value Extraction via Unified Learning Scheme and Dynamic Range Minimization
null
null
null
null
cs.CV cs.CL cs.MM
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
With the prosperity of e-commerce industry, various modalities, e.g., vision and language, are utilized to describe product items. It is an enormous challenge to understand such diversified data, especially via extracting the attribute-value pairs in text sequences with the aid of helpful image regions. Although a series of previous works have been dedicated to this task, there remain seldomly investigated obstacles that hinder further improvements: 1) Parameters from up-stream single-modal pretraining are inadequately applied, without proper jointly fine-tuning in a down-stream multi-modal task. 2) To select descriptive parts of images, a simple late fusion is widely applied, regardless of priori knowledge that language-related information should be encoded into a common linguistic embedding space by stronger encoders. 3) Due to diversity across products, their attribute sets tend to vary greatly, but current approaches predict with an unnecessary maximal range and lead to more potential false positives. To address these issues, we propose in this paper a novel approach to boost multi-modal e-commerce attribute value extraction via unified learning scheme and dynamic range minimization: 1) Firstly, a unified scheme is designed to jointly train a multi-modal task with pretrained single-modal parameters. 2) Secondly, a text-guided information range minimization method is proposed to adaptively encode descriptive parts of each modality into an identical space with a powerful pretrained linguistic model. 3) Moreover, a prototype-guided attribute range minimization method is proposed to first determine the proper attribute set of the current product, and then select prototypes to guide the prediction of the chosen attributes. Experiments on the popular multi-modal e-commerce benchmarks show that our approach achieves superior performance over the other state-of-the-art techniques.
[ { "version": "v1", "created": "Fri, 15 Jul 2022 03:58:04 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 15:16:59 GMT" } ]
2023-04-07T00:00:00
[ [ "Liu", "Mengyin", "" ], [ "Zhu", "Chao", "" ], [ "Gao", "Hongyu", "" ], [ "Gu", "Weibo", "" ], [ "Wang", "Hongfa", "" ], [ "Liu", "Wei", "" ], [ "Yin", "Xu-cheng", "" ] ]
TITLE: Boosting Multi-Modal E-commerce Attribute Value Extraction via Unified Learning Scheme and Dynamic Range Minimization ABSTRACT: With the prosperity of e-commerce industry, various modalities, e.g., vision and language, are utilized to describe product items. It is an enormous challenge to understand such diversified data, especially via extracting the attribute-value pairs in text sequences with the aid of helpful image regions. Although a series of previous works have been dedicated to this task, there remain seldomly investigated obstacles that hinder further improvements: 1) Parameters from up-stream single-modal pretraining are inadequately applied, without proper jointly fine-tuning in a down-stream multi-modal task. 2) To select descriptive parts of images, a simple late fusion is widely applied, regardless of priori knowledge that language-related information should be encoded into a common linguistic embedding space by stronger encoders. 3) Due to diversity across products, their attribute sets tend to vary greatly, but current approaches predict with an unnecessary maximal range and lead to more potential false positives. To address these issues, we propose in this paper a novel approach to boost multi-modal e-commerce attribute value extraction via unified learning scheme and dynamic range minimization: 1) Firstly, a unified scheme is designed to jointly train a multi-modal task with pretrained single-modal parameters. 2) Secondly, a text-guided information range minimization method is proposed to adaptively encode descriptive parts of each modality into an identical space with a powerful pretrained linguistic model. 3) Moreover, a prototype-guided attribute range minimization method is proposed to first determine the proper attribute set of the current product, and then select prototypes to guide the prediction of the chosen attributes. Experiments on the popular multi-modal e-commerce benchmarks show that our approach achieves superior performance over the other state-of-the-art techniques.
no_new_dataset
0.71039
2207.10909
Jinrong Yang
Jinrong Yang, Lin Song, Songtao Liu, Weixin Mao, Zeming Li, Xiaoping Li, Hongbin Sun, Jian Sun, Nanning Zheng
DBQ-SSD: Dynamic Ball Query for Efficient 3D Object Detection
ICLR 2023. Code at https://github.com/yancie-yjr/DBQ-SSD
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Many point-based 3D detectors adopt point-feature sampling strategies to drop some points for efficient inference. These strategies are typically based on fixed and handcrafted rules, making it difficult to handle complicated scenes. Different from them, we propose a Dynamic Ball Query (DBQ) network to adaptively select a subset of input points according to the input features, and assign the feature transform with a suitable receptive field for each selected point. It can be embedded into some state-of-the-art 3D detectors and trained in an end-to-end manner, which significantly reduces the computational cost. Extensive experiments demonstrate that our method can increase the inference speed by 30%-100% on KITTI, Waymo, and ONCE datasets. Specifically, the inference speed of our detector can reach 162 FPS on KITTI scene, and 30 FPS on Waymo and ONCE scenes without performance degradation. Due to skipping the redundant points, some evaluation metrics show significant improvements. Codes will be released at https://github.com/yancie-yjr/DBQ-SSD.
[ { "version": "v1", "created": "Fri, 22 Jul 2022 07:08:42 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 13:19:37 GMT" } ]
2023-04-07T00:00:00
[ [ "Yang", "Jinrong", "" ], [ "Song", "Lin", "" ], [ "Liu", "Songtao", "" ], [ "Mao", "Weixin", "" ], [ "Li", "Zeming", "" ], [ "Li", "Xiaoping", "" ], [ "Sun", "Hongbin", "" ], [ "Sun", "Jian", "" ], [ "Zheng", "Nanning", "" ] ]
TITLE: DBQ-SSD: Dynamic Ball Query for Efficient 3D Object Detection ABSTRACT: Many point-based 3D detectors adopt point-feature sampling strategies to drop some points for efficient inference. These strategies are typically based on fixed and handcrafted rules, making it difficult to handle complicated scenes. Different from them, we propose a Dynamic Ball Query (DBQ) network to adaptively select a subset of input points according to the input features, and assign the feature transform with a suitable receptive field for each selected point. It can be embedded into some state-of-the-art 3D detectors and trained in an end-to-end manner, which significantly reduces the computational cost. Extensive experiments demonstrate that our method can increase the inference speed by 30%-100% on KITTI, Waymo, and ONCE datasets. Specifically, the inference speed of our detector can reach 162 FPS on KITTI scene, and 30 FPS on Waymo and ONCE scenes without performance degradation. Due to skipping the redundant points, some evaluation metrics show significant improvements. Codes will be released at https://github.com/yancie-yjr/DBQ-SSD.
no_new_dataset
0.711825
2207.13253
Yuntong Li
Yuntong Li, Shaowei Wang, Yingying Wang, Jin Li, Yuqiu Qian, Bangzhou Xin, Wei Yang
Fine-grained Private Knowledge Distillation
null
null
null
null
cs.CR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Knowledge distillation has emerged as a scalable and effective way for privacy-preserving machine learning. One remaining drawback is that it consumes privacy in a model-level (i.e., client-level) manner, every distillation query incurs privacy loss of one client's all records. In order to attain fine-grained privacy accountant and improve utility, this work proposes a model-free reverse $k$-NN labeling method towards record-level private knowledge distillation, where each record is employed for labeling at most $k$ queries. Theoretically, we provide bounds of labeling error rate under the centralized/local/shuffle model of differential privacy (w.r.t. the number of records per query, privacy budgets). Experimentally, we demonstrate that it achieves new state-of-the-art accuracy with one order of magnitude lower of privacy loss. Specifically, on the CIFAR-$10$ dataset, it reaches $82.1\%$ test accuracy with centralized privacy budget $1.0$; on the MNIST/SVHN dataset, it reaches $99.1\%$/$95.6\%$ accuracy respectively with budget $0.1$. It is the first time deep learning with differential privacy achieve comparable accuracy with reasonable data privacy protection (i.e., $\exp(\epsilon)\leq 1.5$). Our code is available at https://github.com/liyuntong9/rknn.
[ { "version": "v1", "created": "Wed, 27 Jul 2022 02:32:32 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 13:16:15 GMT" } ]
2023-04-07T00:00:00
[ [ "Li", "Yuntong", "" ], [ "Wang", "Shaowei", "" ], [ "Wang", "Yingying", "" ], [ "Li", "Jin", "" ], [ "Qian", "Yuqiu", "" ], [ "Xin", "Bangzhou", "" ], [ "Yang", "Wei", "" ] ]
TITLE: Fine-grained Private Knowledge Distillation ABSTRACT: Knowledge distillation has emerged as a scalable and effective way for privacy-preserving machine learning. One remaining drawback is that it consumes privacy in a model-level (i.e., client-level) manner, every distillation query incurs privacy loss of one client's all records. In order to attain fine-grained privacy accountant and improve utility, this work proposes a model-free reverse $k$-NN labeling method towards record-level private knowledge distillation, where each record is employed for labeling at most $k$ queries. Theoretically, we provide bounds of labeling error rate under the centralized/local/shuffle model of differential privacy (w.r.t. the number of records per query, privacy budgets). Experimentally, we demonstrate that it achieves new state-of-the-art accuracy with one order of magnitude lower of privacy loss. Specifically, on the CIFAR-$10$ dataset, it reaches $82.1\%$ test accuracy with centralized privacy budget $1.0$; on the MNIST/SVHN dataset, it reaches $99.1\%$/$95.6\%$ accuracy respectively with budget $0.1$. It is the first time deep learning with differential privacy achieve comparable accuracy with reasonable data privacy protection (i.e., $\exp(\epsilon)\leq 1.5$). Our code is available at https://github.com/liyuntong9/rknn.
no_new_dataset
0.70883
2209.06185
Joseph DiPalma
Joseph DiPalma, Lorenzo Torresani, Saeed Hassanpour
HistoPerm: A Permutation-Based View Generation Approach for Improving Histopathologic Feature Representation Learning
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Deep learning has been effective for histology image analysis in digital pathology. However, many current deep learning approaches require large, strongly- or weakly-labeled images and regions of interest, which can be time-consuming and resource-intensive to obtain. To address this challenge, we present HistoPerm, a view generation method for representation learning using joint embedding architectures that enhances representation learning for histology images. HistoPerm permutes augmented views of patches extracted from whole-slide histology images to improve classification performance. We evaluated the effectiveness of HistoPerm on two histology image datasets for Celiac disease and Renal Cell Carcinoma, using three widely used joint embedding architecture-based representation learning methods: BYOL, SimCLR, and VICReg. Our results show that HistoPerm consistently improves patch- and slide-level classification performance in terms of accuracy, F1-score, and AUC. Specifically, for patch-level classification accuracy on the Celiac disease dataset, HistoPerm boosts BYOL and VICReg by 8% and SimCLR by 3%. On the Renal Cell Carcinoma dataset, patch-level classification accuracy is increased by 2% for BYOL and VICReg, and by 1% for SimCLR. In addition, on the Celiac disease dataset, models with HistoPerm outperform the fully-supervised baseline model by 6%, 5%, and 2% for BYOL, SimCLR, and VICReg, respectively. For the Renal Cell Carcinoma dataset, HistoPerm lowers the classification accuracy gap for the models up to 10% relative to the fully-supervised baseline. These findings suggest that HistoPerm can be a valuable tool for improving representation learning of histopathology features when access to labeled data is limited and can lead to whole-slide classification results that are comparable to or superior to fully-supervised methods.
[ { "version": "v1", "created": "Tue, 13 Sep 2022 17:35:08 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 21:33:36 GMT" } ]
2023-04-07T00:00:00
[ [ "DiPalma", "Joseph", "" ], [ "Torresani", "Lorenzo", "" ], [ "Hassanpour", "Saeed", "" ] ]
TITLE: HistoPerm: A Permutation-Based View Generation Approach for Improving Histopathologic Feature Representation Learning ABSTRACT: Deep learning has been effective for histology image analysis in digital pathology. However, many current deep learning approaches require large, strongly- or weakly-labeled images and regions of interest, which can be time-consuming and resource-intensive to obtain. To address this challenge, we present HistoPerm, a view generation method for representation learning using joint embedding architectures that enhances representation learning for histology images. HistoPerm permutes augmented views of patches extracted from whole-slide histology images to improve classification performance. We evaluated the effectiveness of HistoPerm on two histology image datasets for Celiac disease and Renal Cell Carcinoma, using three widely used joint embedding architecture-based representation learning methods: BYOL, SimCLR, and VICReg. Our results show that HistoPerm consistently improves patch- and slide-level classification performance in terms of accuracy, F1-score, and AUC. Specifically, for patch-level classification accuracy on the Celiac disease dataset, HistoPerm boosts BYOL and VICReg by 8% and SimCLR by 3%. On the Renal Cell Carcinoma dataset, patch-level classification accuracy is increased by 2% for BYOL and VICReg, and by 1% for SimCLR. In addition, on the Celiac disease dataset, models with HistoPerm outperform the fully-supervised baseline model by 6%, 5%, and 2% for BYOL, SimCLR, and VICReg, respectively. For the Renal Cell Carcinoma dataset, HistoPerm lowers the classification accuracy gap for the models up to 10% relative to the fully-supervised baseline. These findings suggest that HistoPerm can be a valuable tool for improving representation learning of histopathology features when access to labeled data is limited and can lead to whole-slide classification results that are comparable to or superior to fully-supervised methods.
no_new_dataset
0.711443
2209.07034
Zhanpeng Shao Dr.
Zhanpeng Shao, Wen Zhou, Wuzhen Wang, Jianyu Yang, Youfu Li
A Temporal Densely Connected Recurrent Network for Event-based Human Pose Estimation
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Event camera is an emerging bio-inspired vision sensors that report per-pixel brightness changes asynchronously. It holds noticeable advantage of high dynamic range, high speed response, and low power budget that enable it to best capture local motions in uncontrolled environments. This motivates us to unlock the potential of event cameras for human pose estimation, as the human pose estimation with event cameras is rarely explored. Due to the novel paradigm shift from conventional frame-based cameras, however, event signals in a time interval contain very limited information, as event cameras can only capture the moving body parts and ignores those static body parts, resulting in some parts to be incomplete or even disappeared in the time interval. This paper proposes a novel densely connected recurrent architecture to address the problem of incomplete information. By this recurrent architecture, we can explicitly model not only the sequential but also non-sequential geometric consistency across time steps to accumulate information from previous frames to recover the entire human bodies, achieving a stable and accurate human pose estimation from event data. Moreover, to better evaluate our model, we collect a large scale multimodal event-based dataset that comes with human pose annotations, which is by far the most challenging one to the best of our knowledge. The experimental results on two public datasets and our own dataset demonstrate the effectiveness and strength of our approach. Code can be available online for facilitating the future research.
[ { "version": "v1", "created": "Thu, 15 Sep 2022 04:08:18 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 09:36:18 GMT" }, { "version": "v3", "created": "Thu, 6 Apr 2023 02:24:10 GMT" } ]
2023-04-07T00:00:00
[ [ "Shao", "Zhanpeng", "" ], [ "Zhou", "Wen", "" ], [ "Wang", "Wuzhen", "" ], [ "Yang", "Jianyu", "" ], [ "Li", "Youfu", "" ] ]
TITLE: A Temporal Densely Connected Recurrent Network for Event-based Human Pose Estimation ABSTRACT: Event camera is an emerging bio-inspired vision sensors that report per-pixel brightness changes asynchronously. It holds noticeable advantage of high dynamic range, high speed response, and low power budget that enable it to best capture local motions in uncontrolled environments. This motivates us to unlock the potential of event cameras for human pose estimation, as the human pose estimation with event cameras is rarely explored. Due to the novel paradigm shift from conventional frame-based cameras, however, event signals in a time interval contain very limited information, as event cameras can only capture the moving body parts and ignores those static body parts, resulting in some parts to be incomplete or even disappeared in the time interval. This paper proposes a novel densely connected recurrent architecture to address the problem of incomplete information. By this recurrent architecture, we can explicitly model not only the sequential but also non-sequential geometric consistency across time steps to accumulate information from previous frames to recover the entire human bodies, achieving a stable and accurate human pose estimation from event data. Moreover, to better evaluate our model, we collect a large scale multimodal event-based dataset that comes with human pose annotations, which is by far the most challenging one to the best of our knowledge. The experimental results on two public datasets and our own dataset demonstrate the effectiveness and strength of our approach. Code can be available online for facilitating the future research.
new_dataset
0.61855
2209.15154
Markelle Kelly
Markelle Kelly and Padhraic Smyth
Variable-Based Calibration for Machine Learning Classifiers
null
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
The deployment of machine learning classifiers in high-stakes domains requires well-calibrated confidence scores for model predictions. In this paper we introduce the notion of variable-based calibration to characterize calibration properties of a model with respect to a variable of interest, generalizing traditional score-based metrics such as expected calibration error (ECE). In particular, we find that models with near-perfect ECE can exhibit significant miscalibration as a function of features of the data. We demonstrate this phenomenon both theoretically and in practice on multiple well-known datasets, and show that it can persist after the application of existing calibration methods. To mitigate this issue, we propose strategies for detection, visualization, and quantification of variable-based calibration error. We then examine the limitations of current score-based calibration methods and explore potential modifications. Finally, we discuss the implications of these findings, emphasizing that an understanding of calibration beyond simple aggregate measures is crucial for endeavors such as fairness and model interpretability.
[ { "version": "v1", "created": "Fri, 30 Sep 2022 00:49:31 GMT" }, { "version": "v2", "created": "Fri, 16 Dec 2022 15:01:34 GMT" }, { "version": "v3", "created": "Wed, 5 Apr 2023 21:15:41 GMT" } ]
2023-04-07T00:00:00
[ [ "Kelly", "Markelle", "" ], [ "Smyth", "Padhraic", "" ] ]
TITLE: Variable-Based Calibration for Machine Learning Classifiers ABSTRACT: The deployment of machine learning classifiers in high-stakes domains requires well-calibrated confidence scores for model predictions. In this paper we introduce the notion of variable-based calibration to characterize calibration properties of a model with respect to a variable of interest, generalizing traditional score-based metrics such as expected calibration error (ECE). In particular, we find that models with near-perfect ECE can exhibit significant miscalibration as a function of features of the data. We demonstrate this phenomenon both theoretically and in practice on multiple well-known datasets, and show that it can persist after the application of existing calibration methods. To mitigate this issue, we propose strategies for detection, visualization, and quantification of variable-based calibration error. We then examine the limitations of current score-based calibration methods and explore potential modifications. Finally, we discuss the implications of these findings, emphasizing that an understanding of calibration beyond simple aggregate measures is crucial for endeavors such as fairness and model interpretability.
no_new_dataset
0.709252
2210.03070
Marta R. Costa-Juss\`a
Marta R. Costa-juss\`a, Eric Smith, Christophe Ropers, Daniel Licht, Jean Maillard, Javier Ferrando, Carlos Escolano
Toxicity in Multilingual Machine Translation at Scale
null
null
null
null
cs.CL
http://creativecommons.org/licenses/by-sa/4.0/
Machine Translation systems can produce different types of errors, some of which are characterized as critical or catastrophic due to the specific negative impact that they can have on users. In this paper we focus on one type of critical error: added toxicity. We evaluate and analyze added toxicity when translating a large evaluation dataset (HOLISTICBIAS, over 472k sentences, covering 13 demographic axes) from English into 164 languages. An automatic toxicity evaluation shows that added toxicity across languages varies from 0% to 5%. The output languages with the most added toxicity tend to be low-resource ones, and the demographic axes with the most added toxicity include sexual orientation, gender and sex, and ability. We also perform human evaluation on a subset of 8 translation directions, confirming the prevalence of true added toxicity. We use a measurement of the amount of source contribution to the translation, where a low source contribution implies hallucination, to interpret what causes toxicity. Making use of the input attributions allows us to explain toxicity, because the source contributions significantly correlate with toxicity for 84% of languages studied. Given our findings, our recommendations to reduce added toxicity are to curate training data to avoid mistranslations, mitigate hallucination and check unstable translations.
[ { "version": "v1", "created": "Thu, 6 Oct 2022 17:26:27 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 20:06:42 GMT" } ]
2023-04-07T00:00:00
[ [ "Costa-jussà", "Marta R.", "" ], [ "Smith", "Eric", "" ], [ "Ropers", "Christophe", "" ], [ "Licht", "Daniel", "" ], [ "Maillard", "Jean", "" ], [ "Ferrando", "Javier", "" ], [ "Escolano", "Carlos", "" ] ]
TITLE: Toxicity in Multilingual Machine Translation at Scale ABSTRACT: Machine Translation systems can produce different types of errors, some of which are characterized as critical or catastrophic due to the specific negative impact that they can have on users. In this paper we focus on one type of critical error: added toxicity. We evaluate and analyze added toxicity when translating a large evaluation dataset (HOLISTICBIAS, over 472k sentences, covering 13 demographic axes) from English into 164 languages. An automatic toxicity evaluation shows that added toxicity across languages varies from 0% to 5%. The output languages with the most added toxicity tend to be low-resource ones, and the demographic axes with the most added toxicity include sexual orientation, gender and sex, and ability. We also perform human evaluation on a subset of 8 translation directions, confirming the prevalence of true added toxicity. We use a measurement of the amount of source contribution to the translation, where a low source contribution implies hallucination, to interpret what causes toxicity. Making use of the input attributions allows us to explain toxicity, because the source contributions significantly correlate with toxicity for 84% of languages studied. Given our findings, our recommendations to reduce added toxicity are to curate training data to avoid mistranslations, mitigate hallucination and check unstable translations.
no_new_dataset
0.683802
2210.04787
Junhong Lin
Junhong Lin, Nanfeng Jiang, Zhentao Zhang, Weiling Chen and Tiesong Zhao
LMQFormer: A Laplace-Prior-Guided Mask Query Transformer for Lightweight Snow Removal
11 pages, 13 figures
null
10.1109/TCSVT.2023.3264824
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Snow removal aims to locate snow areas and recover clean images without repairing traces. Unlike the regularity and semitransparency of rain, snow with various patterns and degradations seriously occludes the background. As a result, the state-of-the-art snow removal methods usually retains a large parameter size. In this paper, we propose a lightweight but high-efficient snow removal network called Laplace Mask Query Transformer (LMQFormer). Firstly, we present a Laplace-VQVAE to generate a coarse mask as prior knowledge of snow. Instead of using the mask in dataset, we aim at reducing both the information entropy of snow and the computational cost of recovery. Secondly, we design a Mask Query Transformer (MQFormer) to remove snow with the coarse mask, where we use two parallel encoders and a hybrid decoder to learn extensive snow features under lightweight requirements. Thirdly, we develop a Duplicated Mask Query Attention (DMQA) that converts the coarse mask into a specific number of queries, which constraint the attention areas of MQFormer with reduced parameters. Experimental results in popular datasets have demonstrated the efficiency of our proposed model, which achieves the state-of-the-art snow removal quality with significantly reduced parameters and the lowest running time.
[ { "version": "v1", "created": "Mon, 10 Oct 2022 15:44:06 GMT" }, { "version": "v2", "created": "Tue, 11 Oct 2022 06:48:37 GMT" }, { "version": "v3", "created": "Wed, 12 Oct 2022 07:45:45 GMT" }, { "version": "v4", "created": "Thu, 6 Apr 2023 03:39:27 GMT" } ]
2023-04-07T00:00:00
[ [ "Lin", "Junhong", "" ], [ "Jiang", "Nanfeng", "" ], [ "Zhang", "Zhentao", "" ], [ "Chen", "Weiling", "" ], [ "Zhao", "Tiesong", "" ] ]
TITLE: LMQFormer: A Laplace-Prior-Guided Mask Query Transformer for Lightweight Snow Removal ABSTRACT: Snow removal aims to locate snow areas and recover clean images without repairing traces. Unlike the regularity and semitransparency of rain, snow with various patterns and degradations seriously occludes the background. As a result, the state-of-the-art snow removal methods usually retains a large parameter size. In this paper, we propose a lightweight but high-efficient snow removal network called Laplace Mask Query Transformer (LMQFormer). Firstly, we present a Laplace-VQVAE to generate a coarse mask as prior knowledge of snow. Instead of using the mask in dataset, we aim at reducing both the information entropy of snow and the computational cost of recovery. Secondly, we design a Mask Query Transformer (MQFormer) to remove snow with the coarse mask, where we use two parallel encoders and a hybrid decoder to learn extensive snow features under lightweight requirements. Thirdly, we develop a Duplicated Mask Query Attention (DMQA) that converts the coarse mask into a specific number of queries, which constraint the attention areas of MQFormer with reduced parameters. Experimental results in popular datasets have demonstrated the efficiency of our proposed model, which achieves the state-of-the-art snow removal quality with significantly reduced parameters and the lowest running time.
no_new_dataset
0.709598
2211.02375
Francesca Cairoli
Francesca Cairoli, Nicola Paoletti, Luca Bortolussi
Conformal Quantitative Predictive Monitoring of STL Requirements for Stochastic Processes
null
null
null
null
eess.SY cs.LG cs.LO cs.RO cs.SY
http://creativecommons.org/licenses/by/4.0/
We consider the problem of predictive monitoring (PM), i.e., predicting at runtime the satisfaction of a desired property from the current system's state. Due to its relevance for runtime safety assurance and online control, PM methods need to be efficient to enable timely interventions against predicted violations, while providing correctness guarantees. We introduce \textit{quantitative predictive monitoring (QPM)}, the first PM method to support stochastic processes and rich specifications given in Signal Temporal Logic (STL). Unlike most of the existing PM techniques that predict whether or not some property $\phi$ is satisfied, QPM provides a quantitative measure of satisfaction by predicting the quantitative (aka robust) STL semantics of $\phi$. QPM derives prediction intervals that are highly efficient to compute and with probabilistic guarantees, in that the intervals cover with arbitrary probability the STL robustness values relative to the stochastic evolution of the system. To do so, we take a machine-learning approach and leverage recent advances in conformal inference for quantile regression, thereby avoiding expensive Monte-Carlo simulations at runtime to estimate the intervals. We also show how our monitors can be combined in a compositional manner to handle composite formulas, without retraining the predictors nor sacrificing the guarantees. We demonstrate the effectiveness and scalability of QPM over a benchmark of four discrete-time stochastic processes with varying degrees of complexity.
[ { "version": "v1", "created": "Fri, 4 Nov 2022 11:08:29 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 09:22:28 GMT" } ]
2023-04-07T00:00:00
[ [ "Cairoli", "Francesca", "" ], [ "Paoletti", "Nicola", "" ], [ "Bortolussi", "Luca", "" ] ]
TITLE: Conformal Quantitative Predictive Monitoring of STL Requirements for Stochastic Processes ABSTRACT: We consider the problem of predictive monitoring (PM), i.e., predicting at runtime the satisfaction of a desired property from the current system's state. Due to its relevance for runtime safety assurance and online control, PM methods need to be efficient to enable timely interventions against predicted violations, while providing correctness guarantees. We introduce \textit{quantitative predictive monitoring (QPM)}, the first PM method to support stochastic processes and rich specifications given in Signal Temporal Logic (STL). Unlike most of the existing PM techniques that predict whether or not some property $\phi$ is satisfied, QPM provides a quantitative measure of satisfaction by predicting the quantitative (aka robust) STL semantics of $\phi$. QPM derives prediction intervals that are highly efficient to compute and with probabilistic guarantees, in that the intervals cover with arbitrary probability the STL robustness values relative to the stochastic evolution of the system. To do so, we take a machine-learning approach and leverage recent advances in conformal inference for quantile regression, thereby avoiding expensive Monte-Carlo simulations at runtime to estimate the intervals. We also show how our monitors can be combined in a compositional manner to handle composite formulas, without retraining the predictors nor sacrificing the guarantees. We demonstrate the effectiveness and scalability of QPM over a benchmark of four discrete-time stochastic processes with varying degrees of complexity.
no_new_dataset
0.706361
2211.05992
Debdipta Goswami
Debdipta Goswami
Delay Embedded Echo-State Network: A Predictor for Partially Observed Systems
7 pages, 10 figures
null
null
null
eess.SY cs.LG cs.SY
http://creativecommons.org/licenses/by-nc-sa/4.0/
This paper considers the problem of data-driven prediction of partially observed systems using a recurrent neural network. While neural network based dynamic predictors perform well with full-state training data, prediction with partial observation during training phase poses a significant challenge. Here a predictor for partial observations is developed using an echo-state network (ESN) and time delay embedding of the partially observed state. The proposed method is theoretically justified with Taken's embedding theorem and strong observability of a nonlinear system. The efficacy of the proposed method is demonstrated on three systems: two synthetic datasets from chaotic dynamical systems and a set of real-time traffic data.
[ { "version": "v1", "created": "Fri, 11 Nov 2022 04:13:55 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 22:40:17 GMT" } ]
2023-04-07T00:00:00
[ [ "Goswami", "Debdipta", "" ] ]
TITLE: Delay Embedded Echo-State Network: A Predictor for Partially Observed Systems ABSTRACT: This paper considers the problem of data-driven prediction of partially observed systems using a recurrent neural network. While neural network based dynamic predictors perform well with full-state training data, prediction with partial observation during training phase poses a significant challenge. Here a predictor for partial observations is developed using an echo-state network (ESN) and time delay embedding of the partially observed state. The proposed method is theoretically justified with Taken's embedding theorem and strong observability of a nonlinear system. The efficacy of the proposed method is demonstrated on three systems: two synthetic datasets from chaotic dynamical systems and a set of real-time traffic data.
no_new_dataset
0.709466
2211.09324
Jeongwhan Choi
Jeongwhan Choi, Seoyoung Hong, Noseong Park, Sung-Bae Cho
Blurring-Sharpening Process Models for Collaborative Filtering
Accepted by SIGIR 2023
null
null
null
cs.IR cs.LG
http://creativecommons.org/licenses/by-nc-nd/4.0/
Collaborative filtering is one of the most fundamental topics for recommender systems. Various methods have been proposed for collaborative filtering, ranging from matrix factorization to graph convolutional methods. Being inspired by recent successes of graph filtering-based methods and score-based generative models (SGMs), we present a novel concept of blurring-sharpening process model (BSPM). SGMs and BSPMs share the same processing philosophy that new information can be discovered (e.g., new images are generated in the case of SGMs) while original information is first perturbed and then recovered to its original form. However, SGMs and our BSPMs deal with different types of information, and their optimal perturbation and recovery processes have fundamental discrepancies. Therefore, our BSPMs have different forms from SGMs. In addition, our concept not only theoretically subsumes many existing collaborative filtering models but also outperforms them in terms of Recall and NDCG in the three benchmark datasets, Gowalla, Yelp2018, and Amazon-book. In addition, the processing time of our method is comparable to other fast baselines. Our proposed concept has much potential in the future to be enhanced by designing better blurring (i.e., perturbation) and sharpening (i.e., recovery) processes than what we use in this paper.
[ { "version": "v1", "created": "Thu, 17 Nov 2022 03:59:54 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 13:57:54 GMT" } ]
2023-04-07T00:00:00
[ [ "Choi", "Jeongwhan", "" ], [ "Hong", "Seoyoung", "" ], [ "Park", "Noseong", "" ], [ "Cho", "Sung-Bae", "" ] ]
TITLE: Blurring-Sharpening Process Models for Collaborative Filtering ABSTRACT: Collaborative filtering is one of the most fundamental topics for recommender systems. Various methods have been proposed for collaborative filtering, ranging from matrix factorization to graph convolutional methods. Being inspired by recent successes of graph filtering-based methods and score-based generative models (SGMs), we present a novel concept of blurring-sharpening process model (BSPM). SGMs and BSPMs share the same processing philosophy that new information can be discovered (e.g., new images are generated in the case of SGMs) while original information is first perturbed and then recovered to its original form. However, SGMs and our BSPMs deal with different types of information, and their optimal perturbation and recovery processes have fundamental discrepancies. Therefore, our BSPMs have different forms from SGMs. In addition, our concept not only theoretically subsumes many existing collaborative filtering models but also outperforms them in terms of Recall and NDCG in the three benchmark datasets, Gowalla, Yelp2018, and Amazon-book. In addition, the processing time of our method is comparable to other fast baselines. Our proposed concept has much potential in the future to be enhanced by designing better blurring (i.e., perturbation) and sharpening (i.e., recovery) processes than what we use in this paper.
no_new_dataset
0.709044
2211.10528
Mengmeng Xu
Mengmeng Xu, Yanghao Li, Cheng-Yang Fu, Bernard Ghanem, Tao Xiang, Juan-Manuel Perez-Rua
Where is my Wallet? Modeling Object Proposal Sets for Egocentric Visual Query Localization
We ranked first and second in the VQ2D and VQ3D tasks in the 2nd Ego4D challenge
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper deals with the problem of localizing objects in image and video datasets from visual exemplars. In particular, we focus on the challenging problem of egocentric visual query localization. We first identify grave implicit biases in current query-conditioned model design and visual query datasets. Then, we directly tackle such biases at both frame and object set levels. Concretely, our method solves these issues by expanding limited annotations and dynamically dropping object proposals during training. Additionally, we propose a novel transformer-based module that allows for object-proposal set context to be considered while incorporating query information. We name our module Conditioned Contextual Transformer or CocoFormer. Our experiments show the proposed adaptations improve egocentric query detection, leading to a better visual query localization system in both 2D and 3D configurations. Thus, we are able to improve frame-level detection performance from 26.28% to 31.26 in AP, which correspondingly improves the VQ2D and VQ3D localization scores by significant margins. Our improved context-aware query object detector ranked first and second in the VQ2D and VQ3D tasks in the 2nd Ego4D challenge. In addition to this, we showcase the relevance of our proposed model in the Few-Shot Detection (FSD) task, where we also achieve SOTA results. Our code is available at https://github.com/facebookresearch/vq2d_cvpr.
[ { "version": "v1", "created": "Fri, 18 Nov 2022 22:50:50 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 09:21:18 GMT" } ]
2023-04-07T00:00:00
[ [ "Xu", "Mengmeng", "" ], [ "Li", "Yanghao", "" ], [ "Fu", "Cheng-Yang", "" ], [ "Ghanem", "Bernard", "" ], [ "Xiang", "Tao", "" ], [ "Perez-Rua", "Juan-Manuel", "" ] ]
TITLE: Where is my Wallet? Modeling Object Proposal Sets for Egocentric Visual Query Localization ABSTRACT: This paper deals with the problem of localizing objects in image and video datasets from visual exemplars. In particular, we focus on the challenging problem of egocentric visual query localization. We first identify grave implicit biases in current query-conditioned model design and visual query datasets. Then, we directly tackle such biases at both frame and object set levels. Concretely, our method solves these issues by expanding limited annotations and dynamically dropping object proposals during training. Additionally, we propose a novel transformer-based module that allows for object-proposal set context to be considered while incorporating query information. We name our module Conditioned Contextual Transformer or CocoFormer. Our experiments show the proposed adaptations improve egocentric query detection, leading to a better visual query localization system in both 2D and 3D configurations. Thus, we are able to improve frame-level detection performance from 26.28% to 31.26 in AP, which correspondingly improves the VQ2D and VQ3D localization scores by significant margins. Our improved context-aware query object detector ranked first and second in the VQ2D and VQ3D tasks in the 2nd Ego4D challenge. In addition to this, we showcase the relevance of our proposed model in the Few-Shot Detection (FSD) task, where we also achieve SOTA results. Our code is available at https://github.com/facebookresearch/vq2d_cvpr.
no_new_dataset
0.708584
2211.13222
Zhen Xing
Zhen Xing and Qi Dai and Han Hu and Jingjing Chen and Zuxuan Wu and Yu-Gang Jiang
SVFormer: Semi-supervised Video Transformer for Action Recognition
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Semi-supervised action recognition is a challenging but critical task due to the high cost of video annotations. Existing approaches mainly use convolutional neural networks, yet current revolutionary vision transformer models have been less explored. In this paper, we investigate the use of transformer models under the SSL setting for action recognition. To this end, we introduce SVFormer, which adopts a steady pseudo-labeling framework (ie, EMA-Teacher) to cope with unlabeled video samples. While a wide range of data augmentations have been shown effective for semi-supervised image classification, they generally produce limited results for video recognition. We therefore introduce a novel augmentation strategy, Tube TokenMix, tailored for video data where video clips are mixed via a mask with consistent masked tokens over the temporal axis. In addition, we propose a temporal warping augmentation to cover the complex temporal variation in videos, which stretches selected frames to various temporal durations in the clip. Extensive experiments on three datasets Kinetics-400, UCF-101, and HMDB-51 verify the advantage of SVFormer. In particular, SVFormer outperforms the state-of-the-art by 31.5% with fewer training epochs under the 1% labeling rate of Kinetics-400. Our method can hopefully serve as a strong benchmark and encourage future search on semi-supervised action recognition with Transformer networks.
[ { "version": "v1", "created": "Wed, 23 Nov 2022 18:58:42 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 12:48:30 GMT" } ]
2023-04-07T00:00:00
[ [ "Xing", "Zhen", "" ], [ "Dai", "Qi", "" ], [ "Hu", "Han", "" ], [ "Chen", "Jingjing", "" ], [ "Wu", "Zuxuan", "" ], [ "Jiang", "Yu-Gang", "" ] ]
TITLE: SVFormer: Semi-supervised Video Transformer for Action Recognition ABSTRACT: Semi-supervised action recognition is a challenging but critical task due to the high cost of video annotations. Existing approaches mainly use convolutional neural networks, yet current revolutionary vision transformer models have been less explored. In this paper, we investigate the use of transformer models under the SSL setting for action recognition. To this end, we introduce SVFormer, which adopts a steady pseudo-labeling framework (ie, EMA-Teacher) to cope with unlabeled video samples. While a wide range of data augmentations have been shown effective for semi-supervised image classification, they generally produce limited results for video recognition. We therefore introduce a novel augmentation strategy, Tube TokenMix, tailored for video data where video clips are mixed via a mask with consistent masked tokens over the temporal axis. In addition, we propose a temporal warping augmentation to cover the complex temporal variation in videos, which stretches selected frames to various temporal durations in the clip. Extensive experiments on three datasets Kinetics-400, UCF-101, and HMDB-51 verify the advantage of SVFormer. In particular, SVFormer outperforms the state-of-the-art by 31.5% with fewer training epochs under the 1% labeling rate of Kinetics-400. Our method can hopefully serve as a strong benchmark and encourage future search on semi-supervised action recognition with Transformer networks.
no_new_dataset
0.708036
2211.15654
Songyou Peng
Songyou Peng, Kyle Genova, Chiyu "Max" Jiang, Andrea Tagliasacchi, Marc Pollefeys, Thomas Funkhouser
OpenScene: 3D Scene Understanding with Open Vocabularies
CVPR 2023. Project page: https://pengsongyou.github.io/openscene
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Traditional 3D scene understanding approaches rely on labeled 3D datasets to train a model for a single task with supervision. We propose OpenScene, an alternative approach where a model predicts dense features for 3D scene points that are co-embedded with text and image pixels in CLIP feature space. This zero-shot approach enables task-agnostic training and open-vocabulary queries. For example, to perform SOTA zero-shot 3D semantic segmentation it first infers CLIP features for every 3D point and later classifies them based on similarities to embeddings of arbitrary class labels. More interestingly, it enables a suite of open-vocabulary scene understanding applications that have never been done before. For example, it allows a user to enter an arbitrary text query and then see a heat map indicating which parts of a scene match. Our approach is effective at identifying objects, materials, affordances, activities, and room types in complex 3D scenes, all using a single model trained without any labeled 3D data.
[ { "version": "v1", "created": "Mon, 28 Nov 2022 18:58:36 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 15:35:13 GMT" } ]
2023-04-07T00:00:00
[ [ "Peng", "Songyou", "" ], [ "Genova", "Kyle", "" ], [ "Jiang", "Chiyu \"Max\"", "" ], [ "Tagliasacchi", "Andrea", "" ], [ "Pollefeys", "Marc", "" ], [ "Funkhouser", "Thomas", "" ] ]
TITLE: OpenScene: 3D Scene Understanding with Open Vocabularies ABSTRACT: Traditional 3D scene understanding approaches rely on labeled 3D datasets to train a model for a single task with supervision. We propose OpenScene, an alternative approach where a model predicts dense features for 3D scene points that are co-embedded with text and image pixels in CLIP feature space. This zero-shot approach enables task-agnostic training and open-vocabulary queries. For example, to perform SOTA zero-shot 3D semantic segmentation it first infers CLIP features for every 3D point and later classifies them based on similarities to embeddings of arbitrary class labels. More interestingly, it enables a suite of open-vocabulary scene understanding applications that have never been done before. For example, it allows a user to enter an arbitrary text query and then see a heat map indicating which parts of a scene match. Our approach is effective at identifying objects, materials, affordances, activities, and room types in complex 3D scenes, all using a single model trained without any labeled 3D data.
no_new_dataset
0.711844
2212.14310
Yan Wang
Duowen Chen, Yunhao Bai, Wei Shen, Qingli Li, Lequan Yu, Yan Wang
MagicNet: Semi-Supervised Multi-Organ Segmentation via Magic-Cube Partition and Recovery
Accepted by CVPR 2023
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We propose a novel teacher-student model for semi-supervised multi-organ segmentation. In teacher-student model, data augmentation is usually adopted on unlabeled data to regularize the consistent training between teacher and student. We start from a key perspective that fixed relative locations and variable sizes of different organs can provide distribution information where a multi-organ CT scan is drawn. Thus, we treat the prior anatomy as a strong tool to guide the data augmentation and reduce the mismatch between labeled and unlabeled images for semi-supervised learning. More specifically, we propose a data augmentation strategy based on partition-and-recovery N$^3$ cubes cross- and within- labeled and unlabeled images. Our strategy encourages unlabeled images to learn organ semantics in relative locations from the labeled images (cross-branch) and enhances the learning ability for small organs (within-branch). For within-branch, we further propose to refine the quality of pseudo labels by blending the learned representations from small cubes to incorporate local attributes. Our method is termed as MagicNet, since it treats the CT volume as a magic-cube and N$^3$-cube partition-and-recovery process matches with the rule of playing a magic-cube. Extensive experiments on two public CT multi-organ datasets demonstrate the effectiveness of MagicNet, and noticeably outperforms state-of-the-art semi-supervised medical image segmentation approaches, with +7% DSC improvement on MACT dataset with 10% labeled images. Code is available at https://github.com/DeepMed-Lab-ECNU/MagicNet.
[ { "version": "v1", "created": "Thu, 29 Dec 2022 13:59:38 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 06:16:07 GMT" } ]
2023-04-07T00:00:00
[ [ "Chen", "Duowen", "" ], [ "Bai", "Yunhao", "" ], [ "Shen", "Wei", "" ], [ "Li", "Qingli", "" ], [ "Yu", "Lequan", "" ], [ "Wang", "Yan", "" ] ]
TITLE: MagicNet: Semi-Supervised Multi-Organ Segmentation via Magic-Cube Partition and Recovery ABSTRACT: We propose a novel teacher-student model for semi-supervised multi-organ segmentation. In teacher-student model, data augmentation is usually adopted on unlabeled data to regularize the consistent training between teacher and student. We start from a key perspective that fixed relative locations and variable sizes of different organs can provide distribution information where a multi-organ CT scan is drawn. Thus, we treat the prior anatomy as a strong tool to guide the data augmentation and reduce the mismatch between labeled and unlabeled images for semi-supervised learning. More specifically, we propose a data augmentation strategy based on partition-and-recovery N$^3$ cubes cross- and within- labeled and unlabeled images. Our strategy encourages unlabeled images to learn organ semantics in relative locations from the labeled images (cross-branch) and enhances the learning ability for small organs (within-branch). For within-branch, we further propose to refine the quality of pseudo labels by blending the learned representations from small cubes to incorporate local attributes. Our method is termed as MagicNet, since it treats the CT volume as a magic-cube and N$^3$-cube partition-and-recovery process matches with the rule of playing a magic-cube. Extensive experiments on two public CT multi-organ datasets demonstrate the effectiveness of MagicNet, and noticeably outperforms state-of-the-art semi-supervised medical image segmentation approaches, with +7% DSC improvement on MACT dataset with 10% labeled images. Code is available at https://github.com/DeepMed-Lab-ECNU/MagicNet.
no_new_dataset
0.711794
2301.00508
Fred Buhl
Fred W. Buhl
EmoGator: A New Open Source Vocal Burst Dataset with Baseline Machine Learning Classification Methodologies
12 pages, 4 tables, 2 figures
null
null
null
cs.SD cs.LG eess.AS
http://creativecommons.org/licenses/by/4.0/
Vocal Bursts -- short, non-speech vocalizations that convey emotions, such as laughter, cries, sighs, moans, and groans -- are an often-overlooked aspect of speech emotion recognition, but an important aspect of human vocal communication. One barrier to study of these interesting vocalizations is a lack of large datasets. I am pleased to introduce the EmoGator dataset, which consists of 32,130 samples from 357 speakers, 16.9654 hours of audio; each sample classified into one of 30 distinct emotion categories by the speaker. Several different approaches to construct classifiers to identify emotion categories will be discussed, and directions for future research will be suggested. Data set is available for download from https://github.com/fredbuhl/EmoGator.
[ { "version": "v1", "created": "Mon, 2 Jan 2023 03:02:10 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 15:25:44 GMT" } ]
2023-04-07T00:00:00
[ [ "Buhl", "Fred W.", "" ] ]
TITLE: EmoGator: A New Open Source Vocal Burst Dataset with Baseline Machine Learning Classification Methodologies ABSTRACT: Vocal Bursts -- short, non-speech vocalizations that convey emotions, such as laughter, cries, sighs, moans, and groans -- are an often-overlooked aspect of speech emotion recognition, but an important aspect of human vocal communication. One barrier to study of these interesting vocalizations is a lack of large datasets. I am pleased to introduce the EmoGator dataset, which consists of 32,130 samples from 357 speakers, 16.9654 hours of audio; each sample classified into one of 30 distinct emotion categories by the speaker. Several different approaches to construct classifiers to identify emotion categories will be discussed, and directions for future research will be suggested. Data set is available for download from https://github.com/fredbuhl/EmoGator.
new_dataset
0.714647
2301.04926
Runnan Chen Dr.
Runnan Chen, Youquan Liu, Lingdong Kong, Xinge Zhu, Yuexin Ma, Yikang Li, Yuenan Hou, Yu Qiao, Wenping Wang
CLIP2Scene: Towards Label-efficient 3D Scene Understanding by CLIP
CVPR 2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Contrastive Language-Image Pre-training (CLIP) achieves promising results in 2D zero-shot and few-shot learning. Despite the impressive performance in 2D, applying CLIP to help the learning in 3D scene understanding has yet to be explored. In this paper, we make the first attempt to investigate how CLIP knowledge benefits 3D scene understanding. We propose CLIP2Scene, a simple yet effective framework that transfers CLIP knowledge from 2D image-text pre-trained models to a 3D point cloud network. We show that the pre-trained 3D network yields impressive performance on various downstream tasks, i.e., annotation-free and fine-tuning with labelled data for semantic segmentation. Specifically, built upon CLIP, we design a Semantic-driven Cross-modal Contrastive Learning framework that pre-trains a 3D network via semantic and spatial-temporal consistency regularization. For the former, we first leverage CLIP's text semantics to select the positive and negative point samples and then employ the contrastive loss to train the 3D network. In terms of the latter, we force the consistency between the temporally coherent point cloud features and their corresponding image features. We conduct experiments on SemanticKITTI, nuScenes, and ScanNet. For the first time, our pre-trained network achieves annotation-free 3D semantic segmentation with 20.8% and 25.08% mIoU on nuScenes and ScanNet, respectively. When fine-tuned with 1% or 100% labelled data, our method significantly outperforms other self-supervised methods, with improvements of 8% and 1% mIoU, respectively. Furthermore, we demonstrate the generalizability for handling cross-domain datasets. Code is publicly available https://github.com/runnanchen/CLIP2Scene.
[ { "version": "v1", "created": "Thu, 12 Jan 2023 10:42:39 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 09:56:10 GMT" } ]
2023-04-07T00:00:00
[ [ "Chen", "Runnan", "" ], [ "Liu", "Youquan", "" ], [ "Kong", "Lingdong", "" ], [ "Zhu", "Xinge", "" ], [ "Ma", "Yuexin", "" ], [ "Li", "Yikang", "" ], [ "Hou", "Yuenan", "" ], [ "Qiao", "Yu", "" ], [ "Wang", "Wenping", "" ] ]
TITLE: CLIP2Scene: Towards Label-efficient 3D Scene Understanding by CLIP ABSTRACT: Contrastive Language-Image Pre-training (CLIP) achieves promising results in 2D zero-shot and few-shot learning. Despite the impressive performance in 2D, applying CLIP to help the learning in 3D scene understanding has yet to be explored. In this paper, we make the first attempt to investigate how CLIP knowledge benefits 3D scene understanding. We propose CLIP2Scene, a simple yet effective framework that transfers CLIP knowledge from 2D image-text pre-trained models to a 3D point cloud network. We show that the pre-trained 3D network yields impressive performance on various downstream tasks, i.e., annotation-free and fine-tuning with labelled data for semantic segmentation. Specifically, built upon CLIP, we design a Semantic-driven Cross-modal Contrastive Learning framework that pre-trains a 3D network via semantic and spatial-temporal consistency regularization. For the former, we first leverage CLIP's text semantics to select the positive and negative point samples and then employ the contrastive loss to train the 3D network. In terms of the latter, we force the consistency between the temporally coherent point cloud features and their corresponding image features. We conduct experiments on SemanticKITTI, nuScenes, and ScanNet. For the first time, our pre-trained network achieves annotation-free 3D semantic segmentation with 20.8% and 25.08% mIoU on nuScenes and ScanNet, respectively. When fine-tuned with 1% or 100% labelled data, our method significantly outperforms other self-supervised methods, with improvements of 8% and 1% mIoU, respectively. Furthermore, we demonstrate the generalizability for handling cross-domain datasets. Code is publicly available https://github.com/runnanchen/CLIP2Scene.
no_new_dataset
0.711588
2301.08843
Zhidi Lin
Zhid Lin, Feng Yin and Juan Maro\~nas
Towards Flexibility and Interpretability of Gaussian Process State-Space Model
preprint
null
null
null
cs.LG eess.SP
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The Gaussian process state-space model (GPSSM) has garnered considerable attention over the past decade. However, the standard GP with a preliminary kernel, such as the squared exponential kernel or Mat\'{e}rn kernel, that is commonly used in GPSSM studies, limits the model's representation power and substantially restricts its applicability to complex scenarios. To address this issue, we propose a new class of probabilistic state-space models called TGPSSMs, which leverage a parametric normalizing flow to enrich the GP priors in the standard GPSSM, enabling greater flexibility and expressivity. Additionally, we present a scalable variational inference algorithm that offers a flexible and optimal structure for the variational distribution of latent states. The proposed algorithm is interpretable and computationally efficient due to the sparse GP representation and the bijective nature of normalizing flow. Moreover, we incorporate a constrained optimization framework into the algorithm to enhance the state-space representation capabilities and optimize the hyperparameters, leading to superior learning and inference performance. Experimental results on synthetic and real datasets corroborate that the proposed TGPSSM outperforms several state-of-the-art methods. The accompanying source code is available at \url{https://github.com/zhidilin/TGPSSM}.
[ { "version": "v1", "created": "Sat, 21 Jan 2023 01:26:26 GMT" }, { "version": "v2", "created": "Fri, 31 Mar 2023 03:00:34 GMT" }, { "version": "v3", "created": "Thu, 6 Apr 2023 15:07:11 GMT" } ]
2023-04-07T00:00:00
[ [ "Lin", "Zhid", "" ], [ "Yin", "Feng", "" ], [ "Maroñas", "Juan", "" ] ]
TITLE: Towards Flexibility and Interpretability of Gaussian Process State-Space Model ABSTRACT: The Gaussian process state-space model (GPSSM) has garnered considerable attention over the past decade. However, the standard GP with a preliminary kernel, such as the squared exponential kernel or Mat\'{e}rn kernel, that is commonly used in GPSSM studies, limits the model's representation power and substantially restricts its applicability to complex scenarios. To address this issue, we propose a new class of probabilistic state-space models called TGPSSMs, which leverage a parametric normalizing flow to enrich the GP priors in the standard GPSSM, enabling greater flexibility and expressivity. Additionally, we present a scalable variational inference algorithm that offers a flexible and optimal structure for the variational distribution of latent states. The proposed algorithm is interpretable and computationally efficient due to the sparse GP representation and the bijective nature of normalizing flow. Moreover, we incorporate a constrained optimization framework into the algorithm to enhance the state-space representation capabilities and optimize the hyperparameters, leading to superior learning and inference performance. Experimental results on synthetic and real datasets corroborate that the proposed TGPSSM outperforms several state-of-the-art methods. The accompanying source code is available at \url{https://github.com/zhidilin/TGPSSM}.
no_new_dataset
0.709038
2301.11417
Mert Kilickaya
Mert Kilickaya and Joaquin Vanschoren
Are Labels Needed for Incremental Instance Learning?
Accepted at CVPRW on CLVISION (Oral)
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
In this paper, we learn to classify visual object instances, incrementally and via self-supervision (self-incremental). Our learner observes a single instance at a time, which is then discarded from the dataset. Incremental instance learning is challenging, since longer learning sessions exacerbate forgetfulness, and labeling instances is cumbersome. We overcome these challenges via three contributions: i. We propose VINIL, a self-incremental learner that can learn object instances sequentially, ii. We equip VINIL with self-supervision to by-pass the need for instance labelling, iii. We compare VINIL to label-supervised variants on two large-scale benchmarks, and show that VINIL significantly improves accuracy while reducing forgetfulness.
[ { "version": "v1", "created": "Thu, 26 Jan 2023 21:07:12 GMT" }, { "version": "v2", "created": "Thu, 2 Feb 2023 08:54:56 GMT" }, { "version": "v3", "created": "Sat, 18 Feb 2023 21:27:29 GMT" }, { "version": "v4", "created": "Thu, 6 Apr 2023 10:14:14 GMT" } ]
2023-04-07T00:00:00
[ [ "Kilickaya", "Mert", "" ], [ "Vanschoren", "Joaquin", "" ] ]
TITLE: Are Labels Needed for Incremental Instance Learning? ABSTRACT: In this paper, we learn to classify visual object instances, incrementally and via self-supervision (self-incremental). Our learner observes a single instance at a time, which is then discarded from the dataset. Incremental instance learning is challenging, since longer learning sessions exacerbate forgetfulness, and labeling instances is cumbersome. We overcome these challenges via three contributions: i. We propose VINIL, a self-incremental learner that can learn object instances sequentially, ii. We equip VINIL with self-supervision to by-pass the need for instance labelling, iii. We compare VINIL to label-supervised variants on two large-scale benchmarks, and show that VINIL significantly improves accuracy while reducing forgetfulness.
no_new_dataset
0.709069
2302.11328
Deqiang Li
Deqiang Li, Shicheng Cui, Yun Li, Jia Xu, Fu Xiao and Shouhuai Xu
PAD: Towards Principled Adversarial Malware Detection Against Evasion Attacks
Accepted by IEEE Transactions on Dependable and Secure Computing; To appear
null
10.1109/TDSC.2023.3265665
null
cs.CR cs.LG stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Machine Learning (ML) techniques can facilitate the automation of malicious software (malware for short) detection, but suffer from evasion attacks. Many studies counter such attacks in heuristic manners, lacking theoretical guarantees and defense effectiveness. In this paper, we propose a new adversarial training framework, termed Principled Adversarial Malware Detection (PAD), which offers convergence guarantees for robust optimization methods. PAD lays on a learnable convex measurement that quantifies distribution-wise discrete perturbations to protect malware detectors from adversaries, whereby for smooth detectors, adversarial training can be performed with theoretical treatments. To promote defense effectiveness, we propose a new mixture of attacks to instantiate PAD to enhance deep neural network-based measurements and malware detectors. Experimental results on two Android malware datasets demonstrate: (i) the proposed method significantly outperforms the state-of-the-art defenses; (ii) it can harden ML-based malware detection against 27 evasion attacks with detection accuracies greater than 83.45%, at the price of suffering an accuracy decrease smaller than 2.16% in the absence of attacks; (iii) it matches or outperforms many anti-malware scanners in VirusTotal against realistic adversarial malware.
[ { "version": "v1", "created": "Wed, 22 Feb 2023 12:24:49 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 07:25:14 GMT" } ]
2023-04-07T00:00:00
[ [ "Li", "Deqiang", "" ], [ "Cui", "Shicheng", "" ], [ "Li", "Yun", "" ], [ "Xu", "Jia", "" ], [ "Xiao", "Fu", "" ], [ "Xu", "Shouhuai", "" ] ]
TITLE: PAD: Towards Principled Adversarial Malware Detection Against Evasion Attacks ABSTRACT: Machine Learning (ML) techniques can facilitate the automation of malicious software (malware for short) detection, but suffer from evasion attacks. Many studies counter such attacks in heuristic manners, lacking theoretical guarantees and defense effectiveness. In this paper, we propose a new adversarial training framework, termed Principled Adversarial Malware Detection (PAD), which offers convergence guarantees for robust optimization methods. PAD lays on a learnable convex measurement that quantifies distribution-wise discrete perturbations to protect malware detectors from adversaries, whereby for smooth detectors, adversarial training can be performed with theoretical treatments. To promote defense effectiveness, we propose a new mixture of attacks to instantiate PAD to enhance deep neural network-based measurements and malware detectors. Experimental results on two Android malware datasets demonstrate: (i) the proposed method significantly outperforms the state-of-the-art defenses; (ii) it can harden ML-based malware detection against 27 evasion attacks with detection accuracies greater than 83.45%, at the price of suffering an accuracy decrease smaller than 2.16% in the absence of attacks; (iii) it matches or outperforms many anti-malware scanners in VirusTotal against realistic adversarial malware.
no_new_dataset
0.709202
2304.00202
Xiaojun Jia
Xiaojun Jia, Yong Zhang, Xingxing Wei, Baoyuan Wu, Ke Ma, Jue Wang, and Xiaochun Cao
Improving Fast Adversarial Training with Prior-Guided Knowledge
null
null
null
null
cs.LG cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Fast adversarial training (FAT) is an efficient method to improve robustness. However, the original FAT suffers from catastrophic overfitting, which dramatically and suddenly reduces robustness after a few training epochs. Although various FAT variants have been proposed to prevent overfitting, they require high training costs. In this paper, we investigate the relationship between adversarial example quality and catastrophic overfitting by comparing the training processes of standard adversarial training and FAT. We find that catastrophic overfitting occurs when the attack success rate of adversarial examples becomes worse. Based on this observation, we propose a positive prior-guided adversarial initialization to prevent overfitting by improving adversarial example quality without extra training costs. This initialization is generated by using high-quality adversarial perturbations from the historical training process. We provide theoretical analysis for the proposed initialization and propose a prior-guided regularization method that boosts the smoothness of the loss function. Additionally, we design a prior-guided ensemble FAT method that averages the different model weights of historical models using different decay rates. Our proposed method, called FGSM-PGK, assembles the prior-guided knowledge, i.e., the prior-guided initialization and model weights, acquired during the historical training process. Evaluations of four datasets demonstrate the superiority of the proposed method.
[ { "version": "v1", "created": "Sat, 1 Apr 2023 02:18:12 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 01:06:55 GMT" } ]
2023-04-07T00:00:00
[ [ "Jia", "Xiaojun", "" ], [ "Zhang", "Yong", "" ], [ "Wei", "Xingxing", "" ], [ "Wu", "Baoyuan", "" ], [ "Ma", "Ke", "" ], [ "Wang", "Jue", "" ], [ "Cao", "Xiaochun", "" ] ]
TITLE: Improving Fast Adversarial Training with Prior-Guided Knowledge ABSTRACT: Fast adversarial training (FAT) is an efficient method to improve robustness. However, the original FAT suffers from catastrophic overfitting, which dramatically and suddenly reduces robustness after a few training epochs. Although various FAT variants have been proposed to prevent overfitting, they require high training costs. In this paper, we investigate the relationship between adversarial example quality and catastrophic overfitting by comparing the training processes of standard adversarial training and FAT. We find that catastrophic overfitting occurs when the attack success rate of adversarial examples becomes worse. Based on this observation, we propose a positive prior-guided adversarial initialization to prevent overfitting by improving adversarial example quality without extra training costs. This initialization is generated by using high-quality adversarial perturbations from the historical training process. We provide theoretical analysis for the proposed initialization and propose a prior-guided regularization method that boosts the smoothness of the loss function. Additionally, we design a prior-guided ensemble FAT method that averages the different model weights of historical models using different decay rates. Our proposed method, called FGSM-PGK, assembles the prior-guided knowledge, i.e., the prior-guided initialization and model weights, acquired during the historical training process. Evaluations of four datasets demonstrate the superiority of the proposed method.
no_new_dataset
0.709007
2304.00971
Hanrong Ye
Hanrong Ye, Dan Xu
Joint 2D-3D Multi-Task Learning on Cityscapes-3D: 3D Detection, Segmentation, and Depth Estimation
A supplementary document for "TaskPrompter: Spatial-Channel Multi-Task Prompting for Dense Scene Understanding" accepted by ICLR 2023. Project page: https://github.com/prismformore/Multi-Task-Transformer/tree/main/TaskPrompter
ICLR 2023
null
null
cs.CV cs.LG
http://creativecommons.org/licenses/by-nc-sa/4.0/
This report serves as a supplementary document for TaskPrompter, detailing its implementation on a new joint 2D-3D multi-task learning benchmark based on Cityscapes-3D. TaskPrompter presents an innovative multi-task prompting framework that unifies the learning of (i) task-generic representations, (ii) task-specific representations, and (iii) cross-task interactions, as opposed to previous approaches that separate these learning objectives into different network modules. This unified approach not only reduces the need for meticulous empirical structure design but also significantly enhances the multi-task network's representation learning capability, as the entire model capacity is devoted to optimizing the three objectives simultaneously. TaskPrompter introduces a new multi-task benchmark based on Cityscapes-3D dataset, which requires the multi-task model to concurrently generate predictions for monocular 3D vehicle detection, semantic segmentation, and monocular depth estimation. These tasks are essential for achieving a joint 2D-3D understanding of visual scenes, particularly in the development of autonomous driving systems. On this challenging benchmark, our multi-task model demonstrates strong performance compared to single-task state-of-the-art methods and establishes new state-of-the-art results on the challenging 3D detection and depth estimation tasks.
[ { "version": "v1", "created": "Mon, 3 Apr 2023 13:41:35 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 13:27:21 GMT" }, { "version": "v3", "created": "Thu, 6 Apr 2023 12:58:28 GMT" } ]
2023-04-07T00:00:00
[ [ "Ye", "Hanrong", "" ], [ "Xu", "Dan", "" ] ]
TITLE: Joint 2D-3D Multi-Task Learning on Cityscapes-3D: 3D Detection, Segmentation, and Depth Estimation ABSTRACT: This report serves as a supplementary document for TaskPrompter, detailing its implementation on a new joint 2D-3D multi-task learning benchmark based on Cityscapes-3D. TaskPrompter presents an innovative multi-task prompting framework that unifies the learning of (i) task-generic representations, (ii) task-specific representations, and (iii) cross-task interactions, as opposed to previous approaches that separate these learning objectives into different network modules. This unified approach not only reduces the need for meticulous empirical structure design but also significantly enhances the multi-task network's representation learning capability, as the entire model capacity is devoted to optimizing the three objectives simultaneously. TaskPrompter introduces a new multi-task benchmark based on Cityscapes-3D dataset, which requires the multi-task model to concurrently generate predictions for monocular 3D vehicle detection, semantic segmentation, and monocular depth estimation. These tasks are essential for achieving a joint 2D-3D understanding of visual scenes, particularly in the development of autonomous driving systems. On this challenging benchmark, our multi-task model demonstrates strong performance compared to single-task state-of-the-art methods and establishes new state-of-the-art results on the challenging 3D detection and depth estimation tasks.
new_dataset
0.718761
2304.01218
Weichao Zhou
Yixuan Wang, Weichao Zhou, Jiameng Fan, Zhilu Wang, Jiajun Li, Xin Chen, Chao Huang, Wenchao Li, Qi Zhu
POLAR-Express: Efficient and Precise Formal Reachability Analysis of Neural-Network Controlled Systems
null
null
null
null
eess.SY cs.AI cs.LG cs.SY
http://creativecommons.org/licenses/by/4.0/
Neural networks (NNs) playing the role of controllers have demonstrated impressive empirical performances on challenging control problems. However, the potential adoption of NN controllers in real-life applications also gives rise to a growing concern over the safety of these neural-network controlled systems (NNCSs), especially when used in safety-critical applications. In this work, we present POLAR-Express, an efficient and precise formal reachability analysis tool for verifying the safety of NNCSs. POLAR-Express uses Taylor model arithmetic to propagate Taylor models (TMs) across a neural network layer-by-layer to compute an overapproximation of the neural-network function. It can be applied to analyze any feed-forward neural network with continuous activation functions. We also present a novel approach to propagate TMs more efficiently and precisely across ReLU activation functions. In addition, POLAR-Express provides parallel computation support for the layer-by-layer propagation of TMs, thus significantly improving the efficiency and scalability over its earlier prototype POLAR. Across the comparison with six other state-of-the-art tools on a diverse set of benchmarks, POLAR-Express achieves the best verification efficiency and tightness in the reachable set analysis.
[ { "version": "v1", "created": "Fri, 31 Mar 2023 06:51:36 GMT" }, { "version": "v2", "created": "Wed, 5 Apr 2023 00:30:50 GMT" }, { "version": "v3", "created": "Thu, 6 Apr 2023 02:12:41 GMT" } ]
2023-04-07T00:00:00
[ [ "Wang", "Yixuan", "" ], [ "Zhou", "Weichao", "" ], [ "Fan", "Jiameng", "" ], [ "Wang", "Zhilu", "" ], [ "Li", "Jiajun", "" ], [ "Chen", "Xin", "" ], [ "Huang", "Chao", "" ], [ "Li", "Wenchao", "" ], [ "Zhu", "Qi", "" ] ]
TITLE: POLAR-Express: Efficient and Precise Formal Reachability Analysis of Neural-Network Controlled Systems ABSTRACT: Neural networks (NNs) playing the role of controllers have demonstrated impressive empirical performances on challenging control problems. However, the potential adoption of NN controllers in real-life applications also gives rise to a growing concern over the safety of these neural-network controlled systems (NNCSs), especially when used in safety-critical applications. In this work, we present POLAR-Express, an efficient and precise formal reachability analysis tool for verifying the safety of NNCSs. POLAR-Express uses Taylor model arithmetic to propagate Taylor models (TMs) across a neural network layer-by-layer to compute an overapproximation of the neural-network function. It can be applied to analyze any feed-forward neural network with continuous activation functions. We also present a novel approach to propagate TMs more efficiently and precisely across ReLU activation functions. In addition, POLAR-Express provides parallel computation support for the layer-by-layer propagation of TMs, thus significantly improving the efficiency and scalability over its earlier prototype POLAR. Across the comparison with six other state-of-the-art tools on a diverse set of benchmarks, POLAR-Express achieves the best verification efficiency and tightness in the reachable set analysis.
no_new_dataset
0.709208
2304.01371
Valerio La Gatta
Valerio La Gatta, Luca Luceri, Francesco Fabbri, Emilio Ferrara
The Interconnected Nature of Online Harm and Moderation: Investigating the Cross-Platform Spread of Harmful Content between YouTube and Twitter
14 pages, 8 figures
null
null
null
cs.SI
http://creativecommons.org/licenses/by-nc-nd/4.0/
The proliferation of harmful content shared online poses a threat to online information integrity and the integrity of discussion across platforms. Despite various moderation interventions adopted by social media platforms, researchers and policymakers are calling for holistic solutions. This study explores how a target platform could leverage content that has been deemed harmful on a source platform by investigating the behavior and characteristics of Twitter users responsible for sharing moderated YouTube videos. Using a large-scale dataset of 600M tweets related to the 2020 U.S. election, we find that moderated Youtube videos are extensively shared on Twitter and that users who share these videos also endorse extreme and conspiratorial ideologies. A fraction of these users are eventually suspended by Twitter, but they do not appear to be involved in state-backed information operations. The findings of this study highlight the complex and interconnected nature of harmful cross-platform information diffusion, raising the need for cross-platform moderation strategies.
[ { "version": "v1", "created": "Mon, 3 Apr 2023 20:56:47 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 09:19:11 GMT" } ]
2023-04-07T00:00:00
[ [ "La Gatta", "Valerio", "" ], [ "Luceri", "Luca", "" ], [ "Fabbri", "Francesco", "" ], [ "Ferrara", "Emilio", "" ] ]
TITLE: The Interconnected Nature of Online Harm and Moderation: Investigating the Cross-Platform Spread of Harmful Content between YouTube and Twitter ABSTRACT: The proliferation of harmful content shared online poses a threat to online information integrity and the integrity of discussion across platforms. Despite various moderation interventions adopted by social media platforms, researchers and policymakers are calling for holistic solutions. This study explores how a target platform could leverage content that has been deemed harmful on a source platform by investigating the behavior and characteristics of Twitter users responsible for sharing moderated YouTube videos. Using a large-scale dataset of 600M tweets related to the 2020 U.S. election, we find that moderated Youtube videos are extensively shared on Twitter and that users who share these videos also endorse extreme and conspiratorial ideologies. A fraction of these users are eventually suspended by Twitter, but they do not appear to be involved in state-backed information operations. The findings of this study highlight the complex and interconnected nature of harmful cross-platform information diffusion, raising the need for cross-platform moderation strategies.
no_new_dataset
0.567637
2304.01507
Wenxuan Tu
Wenxuan Tu, Qing Liao, Sihang Zhou, Xin Peng, Chuan Ma, Zhe Liu, Xinwang Liu, Zhiping Cai
RARE: Robust Masked Graph Autoencoder
null
null
null
null
cs.LG cs.AI
http://creativecommons.org/licenses/by/4.0/
Masked graph autoencoder (MGAE) has emerged as a promising self-supervised graph pre-training (SGP) paradigm due to its simplicity and effectiveness. However, existing efforts perform the mask-then-reconstruct operation in the raw data space as is done in computer vision (CV) and natural language processing (NLP) areas, while neglecting the important non-Euclidean property of graph data. As a result, the highly unstable local connection structures largely increase the uncertainty in inferring masked data and decrease the reliability of the exploited self-supervision signals, leading to inferior representations for downstream evaluations. To address this issue, we propose a novel SGP method termed Robust mAsked gRaph autoEncoder (RARE) to improve the certainty in inferring masked data and the reliability of the self-supervision mechanism by further masking and reconstructing node samples in the high-order latent feature space. Through both theoretical and empirical analyses, we have discovered that performing a joint mask-then-reconstruct strategy in both latent feature and raw data spaces could yield improved stability and performance. To this end, we elaborately design a masked latent feature completion scheme, which predicts latent features of masked nodes under the guidance of high-order sample correlations that are hard to be observed from the raw data perspective. Specifically, we first adopt a latent feature predictor to predict the masked latent features from the visible ones. Next, we encode the raw data of masked samples with a momentum graph encoder and subsequently employ the resulting representations to improve predicted results through latent feature matching. Extensive experiments on seventeen datasets have demonstrated the effectiveness and robustness of RARE against state-of-the-art (SOTA) competitors across three downstream tasks.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 03:35:29 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 15:37:21 GMT" } ]
2023-04-07T00:00:00
[ [ "Tu", "Wenxuan", "" ], [ "Liao", "Qing", "" ], [ "Zhou", "Sihang", "" ], [ "Peng", "Xin", "" ], [ "Ma", "Chuan", "" ], [ "Liu", "Zhe", "" ], [ "Liu", "Xinwang", "" ], [ "Cai", "Zhiping", "" ] ]
TITLE: RARE: Robust Masked Graph Autoencoder ABSTRACT: Masked graph autoencoder (MGAE) has emerged as a promising self-supervised graph pre-training (SGP) paradigm due to its simplicity and effectiveness. However, existing efforts perform the mask-then-reconstruct operation in the raw data space as is done in computer vision (CV) and natural language processing (NLP) areas, while neglecting the important non-Euclidean property of graph data. As a result, the highly unstable local connection structures largely increase the uncertainty in inferring masked data and decrease the reliability of the exploited self-supervision signals, leading to inferior representations for downstream evaluations. To address this issue, we propose a novel SGP method termed Robust mAsked gRaph autoEncoder (RARE) to improve the certainty in inferring masked data and the reliability of the self-supervision mechanism by further masking and reconstructing node samples in the high-order latent feature space. Through both theoretical and empirical analyses, we have discovered that performing a joint mask-then-reconstruct strategy in both latent feature and raw data spaces could yield improved stability and performance. To this end, we elaborately design a masked latent feature completion scheme, which predicts latent features of masked nodes under the guidance of high-order sample correlations that are hard to be observed from the raw data perspective. Specifically, we first adopt a latent feature predictor to predict the masked latent features from the visible ones. Next, we encode the raw data of masked samples with a momentum graph encoder and subsequently employ the resulting representations to improve predicted results through latent feature matching. Extensive experiments on seventeen datasets have demonstrated the effectiveness and robustness of RARE against state-of-the-art (SOTA) competitors across three downstream tasks.
no_new_dataset
0.711067
2304.01554
Ashish Sinha
Ashish Sinha, Jonghyun Choi
MEnsA: Mix-up Ensemble Average for Unsupervised Multi Target Domain Adaptation on 3D Point Clouds
Accepted at Workshop on Learning with Limited Labelled Data (CVPR 2023)
null
null
null
cs.CV cs.AI
http://creativecommons.org/licenses/by/4.0/
Unsupervised domain adaptation (UDA) addresses the problem of distribution shift between the unlabelled target domain and labelled source domain. While the single target domain adaptation (STDA) is well studied in the literature for both 2D and 3D vision tasks, multi-target domain adaptation (MTDA) is barely explored for 3D data despite its wide real-world applications such as autonomous driving systems for various geographical and climatic conditions. We establish an MTDA baseline for 3D point cloud data by proposing to mix the feature representations from all domains together to achieve better domain adaptation performance by an ensemble average, which we call Mixup Ensemble Average or MEnsA. With the mixed representation, we use a domain classifier to improve at distinguishing the feature representations of source domain from those of target domains in a shared latent space. In empirical validations on the challenging PointDA-10 dataset, we showcase a clear benefit of our simple method over previous unsupervised STDA and MTDA methods by large margins (up to 17.10% and 4.76% on averaged over all domain shifts).
[ { "version": "v1", "created": "Tue, 4 Apr 2023 06:13:33 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 00:14:29 GMT" } ]
2023-04-07T00:00:00
[ [ "Sinha", "Ashish", "" ], [ "Choi", "Jonghyun", "" ] ]
TITLE: MEnsA: Mix-up Ensemble Average for Unsupervised Multi Target Domain Adaptation on 3D Point Clouds ABSTRACT: Unsupervised domain adaptation (UDA) addresses the problem of distribution shift between the unlabelled target domain and labelled source domain. While the single target domain adaptation (STDA) is well studied in the literature for both 2D and 3D vision tasks, multi-target domain adaptation (MTDA) is barely explored for 3D data despite its wide real-world applications such as autonomous driving systems for various geographical and climatic conditions. We establish an MTDA baseline for 3D point cloud data by proposing to mix the feature representations from all domains together to achieve better domain adaptation performance by an ensemble average, which we call Mixup Ensemble Average or MEnsA. With the mixed representation, we use a domain classifier to improve at distinguishing the feature representations of source domain from those of target domains in a shared latent space. In empirical validations on the challenging PointDA-10 dataset, we showcase a clear benefit of our simple method over previous unsupervised STDA and MTDA methods by large margins (up to 17.10% and 4.76% on averaged over all domain shifts).
no_new_dataset
0.711055
2304.02413
Shuanghong Shen
Shuanghong Shen, Enhong Chen, Bihan Xu, Qi Liu, Zhenya Huang, Linbo Zhu, Yu Su
Quiz-based Knowledge Tracing
12 pages
null
null
null
cs.AI cs.HC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Knowledge tracing (KT) aims to assess individuals' evolving knowledge states according to their learning interactions with different exercises in online learning systems (OIS), which is critical in supporting decision-making for subsequent intelligent services, such as personalized learning source recommendation. Existing researchers have broadly studied KT and developed many effective methods. However, most of them assume that students' historical interactions are uniformly distributed in a continuous sequence, ignoring the fact that actual interaction sequences are organized based on a series of quizzes with clear boundaries, where interactions within a quiz are consecutively completed, but interactions across different quizzes are discrete and may be spaced over days. In this paper, we present the Quiz-based Knowledge Tracing (QKT) model to monitor students' knowledge states according to their quiz-based learning interactions. Specifically, as students' interactions within a quiz are continuous and have the same or similar knowledge concepts, we design the adjacent gate followed by a global average pooling layer to capture the intra-quiz short-term knowledge influence. Then, as various quizzes tend to focus on different knowledge concepts, we respectively measure the inter-quiz knowledge substitution by the gated recurrent unit and the inter-quiz knowledge complementarity by the self-attentive encoder with a novel recency-aware attention mechanism. Finally, we integrate the inter-quiz long-term knowledge substitution and complementarity across different quizzes to output students' evolving knowledge states. Extensive experimental results on three public real-world datasets demonstrate that QKT achieves state-of-the-art performance compared to existing methods. Further analyses confirm that QKT is promising in designing more effective quizzes.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 12:48:42 GMT" }, { "version": "v2", "created": "Thu, 6 Apr 2023 12:15:41 GMT" } ]
2023-04-07T00:00:00
[ [ "Shen", "Shuanghong", "" ], [ "Chen", "Enhong", "" ], [ "Xu", "Bihan", "" ], [ "Liu", "Qi", "" ], [ "Huang", "Zhenya", "" ], [ "Zhu", "Linbo", "" ], [ "Su", "Yu", "" ] ]
TITLE: Quiz-based Knowledge Tracing ABSTRACT: Knowledge tracing (KT) aims to assess individuals' evolving knowledge states according to their learning interactions with different exercises in online learning systems (OIS), which is critical in supporting decision-making for subsequent intelligent services, such as personalized learning source recommendation. Existing researchers have broadly studied KT and developed many effective methods. However, most of them assume that students' historical interactions are uniformly distributed in a continuous sequence, ignoring the fact that actual interaction sequences are organized based on a series of quizzes with clear boundaries, where interactions within a quiz are consecutively completed, but interactions across different quizzes are discrete and may be spaced over days. In this paper, we present the Quiz-based Knowledge Tracing (QKT) model to monitor students' knowledge states according to their quiz-based learning interactions. Specifically, as students' interactions within a quiz are continuous and have the same or similar knowledge concepts, we design the adjacent gate followed by a global average pooling layer to capture the intra-quiz short-term knowledge influence. Then, as various quizzes tend to focus on different knowledge concepts, we respectively measure the inter-quiz knowledge substitution by the gated recurrent unit and the inter-quiz knowledge complementarity by the self-attentive encoder with a novel recency-aware attention mechanism. Finally, we integrate the inter-quiz long-term knowledge substitution and complementarity across different quizzes to output students' evolving knowledge states. Extensive experimental results on three public real-world datasets demonstrate that QKT achieves state-of-the-art performance compared to existing methods. Further analyses confirm that QKT is promising in designing more effective quizzes.
no_new_dataset
0.709038
2304.02653
Neelesh Mungoli
Neelesh Mungoli
Adaptive Ensemble Learning: Boosting Model Performance through Intelligent Feature Fusion in Deep Neural Networks
null
null
null
null
cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper, we present an Adaptive Ensemble Learning framework that aims to boost the performance of deep neural networks by intelligently fusing features through ensemble learning techniques. The proposed framework integrates ensemble learning strategies with deep learning architectures to create a more robust and adaptable model capable of handling complex tasks across various domains. By leveraging intelligent feature fusion methods, the Adaptive Ensemble Learning framework generates more discriminative and effective feature representations, leading to improved model performance and generalization capabilities. We conducted extensive experiments and evaluations on several benchmark datasets, including image classification, object detection, natural language processing, and graph-based learning tasks. The results demonstrate that the proposed framework consistently outperforms baseline models and traditional feature fusion techniques, highlighting its effectiveness in enhancing deep learning models' performance. Furthermore, we provide insights into the impact of intelligent feature fusion on model performance and discuss the potential applications of the Adaptive Ensemble Learning framework in real-world scenarios. The paper also explores the design and implementation of adaptive ensemble models, ensemble training strategies, and meta-learning techniques, which contribute to the framework's versatility and adaptability. In conclusion, the Adaptive Ensemble Learning framework represents a significant advancement in the field of feature fusion and ensemble learning for deep neural networks, with the potential to transform a wide range of applications across multiple domains.
[ { "version": "v1", "created": "Tue, 4 Apr 2023 21:49:49 GMT" } ]
2023-04-07T00:00:00
[ [ "Mungoli", "Neelesh", "" ] ]
TITLE: Adaptive Ensemble Learning: Boosting Model Performance through Intelligent Feature Fusion in Deep Neural Networks ABSTRACT: In this paper, we present an Adaptive Ensemble Learning framework that aims to boost the performance of deep neural networks by intelligently fusing features through ensemble learning techniques. The proposed framework integrates ensemble learning strategies with deep learning architectures to create a more robust and adaptable model capable of handling complex tasks across various domains. By leveraging intelligent feature fusion methods, the Adaptive Ensemble Learning framework generates more discriminative and effective feature representations, leading to improved model performance and generalization capabilities. We conducted extensive experiments and evaluations on several benchmark datasets, including image classification, object detection, natural language processing, and graph-based learning tasks. The results demonstrate that the proposed framework consistently outperforms baseline models and traditional feature fusion techniques, highlighting its effectiveness in enhancing deep learning models' performance. Furthermore, we provide insights into the impact of intelligent feature fusion on model performance and discuss the potential applications of the Adaptive Ensemble Learning framework in real-world scenarios. The paper also explores the design and implementation of adaptive ensemble models, ensemble training strategies, and meta-learning techniques, which contribute to the framework's versatility and adaptability. In conclusion, the Adaptive Ensemble Learning framework represents a significant advancement in the field of feature fusion and ensemble learning for deep neural networks, with the potential to transform a wide range of applications across multiple domains.
no_new_dataset
0.709384
2304.02686
Javier Cabello S\'anchez
Mariano Rodr\'iguez-Arias Fern\'andez, Javier Cabello S\'anchez, Juan Antonio Fern\'andez Torvisco
Chebyshev approximation of exponential data
10 pages
null
null
null
math.NA cs.NA
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper we present an algorithm to fit data via exponentials when the error is measured using the max-norm. We prove the necesssary results to show that the algorithm will converge to the best approximation no matter the dataset.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 18:29:08 GMT" } ]
2023-04-07T00:00:00
[ [ "Fernández", "Mariano Rodríguez-Arias", "" ], [ "Sánchez", "Javier Cabello", "" ], [ "Torvisco", "Juan Antonio Fernández", "" ] ]
TITLE: Chebyshev approximation of exponential data ABSTRACT: In this paper we present an algorithm to fit data via exponentials when the error is measured using the max-norm. We prove the necesssary results to show that the algorithm will converge to the best approximation no matter the dataset.
no_new_dataset
0.708452
2304.02693
Binghui Wang
Wenjie Qu, Youqi Li, Binghui Wang
A Certified Radius-Guided Attack Framework to Image Segmentation Models
Accepted by EuroSP 2023
null
null
null
cs.CV cs.CR cs.LG
http://creativecommons.org/publicdomain/zero/1.0/
Image segmentation is an important problem in many safety-critical applications. Recent studies show that modern image segmentation models are vulnerable to adversarial perturbations, while existing attack methods mainly follow the idea of attacking image classification models. We argue that image segmentation and classification have inherent differences, and design an attack framework specially for image segmentation models. Our attack framework is inspired by certified radius, which was originally used by defenders to defend against adversarial perturbations to classification models. We are the first, from the attacker perspective, to leverage the properties of certified radius and propose a certified radius guided attack framework against image segmentation models. Specifically, we first adapt randomized smoothing, the state-of-the-art certification method for classification models, to derive the pixel's certified radius. We then focus more on disrupting pixels with relatively smaller certified radii and design a pixel-wise certified radius guided loss, when plugged into any existing white-box attack, yields our certified radius-guided white-box attack. Next, we propose the first black-box attack to image segmentation models via bandit. We design a novel gradient estimator, based on bandit feedback, which is query-efficient and provably unbiased and stable. We use this gradient estimator to design a projected bandit gradient descent (PBGD) attack, as well as a certified radius-guided PBGD (CR-PBGD) attack. We prove our PBGD and CR-PBGD attacks can achieve asymptotically optimal attack performance with an optimal rate. We evaluate our certified-radius guided white-box and black-box attacks on multiple modern image segmentation models and datasets. Our results validate the effectiveness of our certified radius-guided attack framework.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 18:43:15 GMT" } ]
2023-04-07T00:00:00
[ [ "Qu", "Wenjie", "" ], [ "Li", "Youqi", "" ], [ "Wang", "Binghui", "" ] ]
TITLE: A Certified Radius-Guided Attack Framework to Image Segmentation Models ABSTRACT: Image segmentation is an important problem in many safety-critical applications. Recent studies show that modern image segmentation models are vulnerable to adversarial perturbations, while existing attack methods mainly follow the idea of attacking image classification models. We argue that image segmentation and classification have inherent differences, and design an attack framework specially for image segmentation models. Our attack framework is inspired by certified radius, which was originally used by defenders to defend against adversarial perturbations to classification models. We are the first, from the attacker perspective, to leverage the properties of certified radius and propose a certified radius guided attack framework against image segmentation models. Specifically, we first adapt randomized smoothing, the state-of-the-art certification method for classification models, to derive the pixel's certified radius. We then focus more on disrupting pixels with relatively smaller certified radii and design a pixel-wise certified radius guided loss, when plugged into any existing white-box attack, yields our certified radius-guided white-box attack. Next, we propose the first black-box attack to image segmentation models via bandit. We design a novel gradient estimator, based on bandit feedback, which is query-efficient and provably unbiased and stable. We use this gradient estimator to design a projected bandit gradient descent (PBGD) attack, as well as a certified radius-guided PBGD (CR-PBGD) attack. We prove our PBGD and CR-PBGD attacks can achieve asymptotically optimal attack performance with an optimal rate. We evaluate our certified-radius guided white-box and black-box attacks on multiple modern image segmentation models and datasets. Our results validate the effectiveness of our certified radius-guided attack framework.
no_new_dataset
0.709636
2304.02695
Zhangyi Cheng
Zhangyi Cheng, Xiang Zhang, Lei Yu, Jianzhuang Liu, Wen Yang, Gui-Song Xia
Recovering Continuous Scene Dynamics from A Single Blurry Image with Events
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper aims at demystifying a single motion-blurred image with events and revealing temporally continuous scene dynamics encrypted behind motion blurs. To achieve this end, an Implicit Video Function (IVF) is learned to represent a single motion blurred image with concurrent events, enabling the latent sharp image restoration of arbitrary timestamps in the range of imaging exposures. Specifically, a dual attention transformer is proposed to efficiently leverage merits from both modalities, i.e., the high temporal resolution of event features and the smoothness of image features, alleviating temporal ambiguities while suppressing the event noise. The proposed network is trained only with the supervision of ground-truth images of limited referenced timestamps. Motion- and texture-guided supervisions are employed simultaneously to enhance restorations of the non-referenced timestamps and improve the overall sharpness. Experiments on synthetic, semi-synthetic, and real-world datasets demonstrate that our proposed method outperforms state-of-the-art methods by a large margin in terms of both objective PSNR and SSIM measurements and subjective evaluations.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 18:44:17 GMT" } ]
2023-04-07T00:00:00
[ [ "Cheng", "Zhangyi", "" ], [ "Zhang", "Xiang", "" ], [ "Yu", "Lei", "" ], [ "Liu", "Jianzhuang", "" ], [ "Yang", "Wen", "" ], [ "Xia", "Gui-Song", "" ] ]
TITLE: Recovering Continuous Scene Dynamics from A Single Blurry Image with Events ABSTRACT: This paper aims at demystifying a single motion-blurred image with events and revealing temporally continuous scene dynamics encrypted behind motion blurs. To achieve this end, an Implicit Video Function (IVF) is learned to represent a single motion blurred image with concurrent events, enabling the latent sharp image restoration of arbitrary timestamps in the range of imaging exposures. Specifically, a dual attention transformer is proposed to efficiently leverage merits from both modalities, i.e., the high temporal resolution of event features and the smoothness of image features, alleviating temporal ambiguities while suppressing the event noise. The proposed network is trained only with the supervision of ground-truth images of limited referenced timestamps. Motion- and texture-guided supervisions are employed simultaneously to enhance restorations of the non-referenced timestamps and improve the overall sharpness. Experiments on synthetic, semi-synthetic, and real-world datasets demonstrate that our proposed method outperforms state-of-the-art methods by a large margin in terms of both objective PSNR and SSIM measurements and subjective evaluations.
no_new_dataset
0.709818
2304.02704
Bharat Joshi
Weihan Wang, Bharat Joshi, Nathaniel Burgdorfer, Konstantinos Batsos, Alberto Quattrini Li, Philippos Mordohai, Ioannis Rekleitis
Real-Time Dense 3D Mapping of Underwater Environments
null
null
null
null
cs.CV cs.RO
http://creativecommons.org/licenses/by-nc-sa/4.0/
This paper addresses real-time dense 3D reconstruction for a resource-constrained Autonomous Underwater Vehicle (AUV). Underwater vision-guided operations are among the most challenging as they combine 3D motion in the presence of external forces, limited visibility, and absence of global positioning. Obstacle avoidance and effective path planning require online dense reconstructions of the environment. Autonomous operation is central to environmental monitoring, marine archaeology, resource utilization, and underwater cave exploration. To address this problem, we propose to use SVIn2, a robust VIO method, together with a real-time 3D reconstruction pipeline. We provide extensive evaluation on four challenging underwater datasets. Our pipeline produces comparable reconstruction with that of COLMAP, the state-of-the-art offline 3D reconstruction method, at high frame rates on a single CPU.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 18:57:35 GMT" } ]
2023-04-07T00:00:00
[ [ "Wang", "Weihan", "" ], [ "Joshi", "Bharat", "" ], [ "Burgdorfer", "Nathaniel", "" ], [ "Batsos", "Konstantinos", "" ], [ "Li", "Alberto Quattrini", "" ], [ "Mordohai", "Philippos", "" ], [ "Rekleitis", "Ioannis", "" ] ]
TITLE: Real-Time Dense 3D Mapping of Underwater Environments ABSTRACT: This paper addresses real-time dense 3D reconstruction for a resource-constrained Autonomous Underwater Vehicle (AUV). Underwater vision-guided operations are among the most challenging as they combine 3D motion in the presence of external forces, limited visibility, and absence of global positioning. Obstacle avoidance and effective path planning require online dense reconstructions of the environment. Autonomous operation is central to environmental monitoring, marine archaeology, resource utilization, and underwater cave exploration. To address this problem, we propose to use SVIn2, a robust VIO method, together with a real-time 3D reconstruction pipeline. We provide extensive evaluation on four challenging underwater datasets. Our pipeline produces comparable reconstruction with that of COLMAP, the state-of-the-art offline 3D reconstruction method, at high frame rates on a single CPU.
no_new_dataset
0.709252
2304.02713
Sohini Roychowdhury
Sohini Roychowdhury
NUMSnet: Nested-U Multi-class Segmentation network for 3D Medical Image Stacks
15 pages, 10 pages, 8 tables
null
null
null
eess.IV cs.CV
http://creativecommons.org/licenses/by-sa/4.0/
Semantic segmentation for medical 3D image stacks enables accurate volumetric reconstructions, computer-aided diagnostics and follow up treatment planning. In this work, we present a novel variant of the Unet model called the NUMSnet that transmits pixel neighborhood features across scans through nested layers to achieve accurate multi-class semantic segmentations with minimal training data. We analyze the semantic segmentation performance of the NUMSnet model in comparison with several Unet model variants to segment 3-7 regions of interest using only 10% of images for training per Lung-CT and Heart-CT volumetric image stacks. The proposed NUMSnet model achieves up to 20% improvement in segmentation recall with 4-9% improvement in Dice scores for Lung-CT stacks and 2.5-10% improvement in Dice scores for Heart-CT stacks when compared to the Unet++ model. The NUMSnet model needs to be trained by ordered images around the central scan of each volumetric stack. Propagation of image feature information from the 6 nested layers of the Unet++ model are found to have better computation and segmentation performances than propagation of all up-sampling layers in a Unet++ model. The NUMSnet model achieves comparable segmentation performances to existing works, while being trained on as low as 5\% of the training images. Also, transfer learning allows faster convergence of the NUMSnet model for multi-class semantic segmentation from pathology in Lung-CT images to cardiac segmentations in Heart-CT stacks. Thus, the proposed model can standardize multi-class semantic segmentation on a variety of volumetric image stacks with minimal training dataset. This can significantly reduce the cost, time and inter-observer variabilities associated with computer-aided detections and treatment.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 19:16:29 GMT" } ]
2023-04-07T00:00:00
[ [ "Roychowdhury", "Sohini", "" ] ]
TITLE: NUMSnet: Nested-U Multi-class Segmentation network for 3D Medical Image Stacks ABSTRACT: Semantic segmentation for medical 3D image stacks enables accurate volumetric reconstructions, computer-aided diagnostics and follow up treatment planning. In this work, we present a novel variant of the Unet model called the NUMSnet that transmits pixel neighborhood features across scans through nested layers to achieve accurate multi-class semantic segmentations with minimal training data. We analyze the semantic segmentation performance of the NUMSnet model in comparison with several Unet model variants to segment 3-7 regions of interest using only 10% of images for training per Lung-CT and Heart-CT volumetric image stacks. The proposed NUMSnet model achieves up to 20% improvement in segmentation recall with 4-9% improvement in Dice scores for Lung-CT stacks and 2.5-10% improvement in Dice scores for Heart-CT stacks when compared to the Unet++ model. The NUMSnet model needs to be trained by ordered images around the central scan of each volumetric stack. Propagation of image feature information from the 6 nested layers of the Unet++ model are found to have better computation and segmentation performances than propagation of all up-sampling layers in a Unet++ model. The NUMSnet model achieves comparable segmentation performances to existing works, while being trained on as low as 5\% of the training images. Also, transfer learning allows faster convergence of the NUMSnet model for multi-class semantic segmentation from pathology in Lung-CT images to cardiac segmentations in Heart-CT stacks. Thus, the proposed model can standardize multi-class semantic segmentation on a variety of volumetric image stacks with minimal training dataset. This can significantly reduce the cost, time and inter-observer variabilities associated with computer-aided detections and treatment.
no_new_dataset
0.7138
2304.02715
Pu Li
Pu Li, Xiaobai Liu
Learning Knowledge-Rich Sequential Model for Planar Homography Estimation in Aerial Video
Accepted by 2020 25th International Conference on Pattern Recognition (ICPR) 2021
null
10.1109/ICPR48806.2021.9413272
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
This paper presents an unsupervised approach that leverages raw aerial videos to learn to estimate planar homographic transformation between consecutive video frames. Previous learning-based estimators work on pairs of images to estimate their planar homographic transformations but suffer from severe over-fitting issues, especially when applying over aerial videos. To address this concern, we develop a sequential estimator that directly processes a sequence of video frames and estimates their pairwise planar homographic transformations in batches. We also incorporate a set of spatial-temporal knowledge to regularize the learning of such a sequence-to-sequence model. We collect a set of challenging aerial videos and compare the proposed method to the alternative algorithms. Empirical studies suggest that our sequential model achieves significant improvement over alternative image-based methods and the knowledge-rich regularization further boosts our system performance. Our codes and dataset could be found at https://github.com/Paul-LiPu/DeepVideoHomography
[ { "version": "v1", "created": "Wed, 5 Apr 2023 19:28:58 GMT" } ]
2023-04-07T00:00:00
[ [ "Li", "Pu", "" ], [ "Liu", "Xiaobai", "" ] ]
TITLE: Learning Knowledge-Rich Sequential Model for Planar Homography Estimation in Aerial Video ABSTRACT: This paper presents an unsupervised approach that leverages raw aerial videos to learn to estimate planar homographic transformation between consecutive video frames. Previous learning-based estimators work on pairs of images to estimate their planar homographic transformations but suffer from severe over-fitting issues, especially when applying over aerial videos. To address this concern, we develop a sequential estimator that directly processes a sequence of video frames and estimates their pairwise planar homographic transformations in batches. We also incorporate a set of spatial-temporal knowledge to regularize the learning of such a sequence-to-sequence model. We collect a set of challenging aerial videos and compare the proposed method to the alternative algorithms. Empirical studies suggest that our sequential model achieves significant improvement over alternative image-based methods and the knowledge-rich regularization further boosts our system performance. Our codes and dataset could be found at https://github.com/Paul-LiPu/DeepVideoHomography
no_new_dataset
0.703384
2304.02720
Zheyuan Zhang
Zheyuan Zhang, Bin Wang, Lanhong Yao, Ugur Demir, Debesh Jha, Ismail Baris Turkbey, Boqing Gong, Ulas Bagci
Domain Generalization with Adversarial Intensity Attack for Medical Image Segmentation
Code is available upon publication
null
null
null
eess.IV cs.CR cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Most statistical learning algorithms rely on an over-simplified assumption, that is, the train and test data are independent and identically distributed. In real-world scenarios, however, it is common for models to encounter data from new and different domains to which they were not exposed to during training. This is often the case in medical imaging applications due to differences in acquisition devices, imaging protocols, and patient characteristics. To address this problem, domain generalization (DG) is a promising direction as it enables models to handle data from previously unseen domains by learning domain-invariant features robust to variations across different domains. To this end, we introduce a novel DG method called Adversarial Intensity Attack (AdverIN), which leverages adversarial training to generate training data with an infinite number of styles and increase data diversity while preserving essential content information. We conduct extensive evaluation experiments on various multi-domain segmentation datasets, including 2D retinal fundus optic disc/cup and 3D prostate MRI. Our results demonstrate that AdverIN significantly improves the generalization ability of the segmentation models, achieving significant improvement on these challenging datasets. Code is available upon publication.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 19:40:51 GMT" } ]
2023-04-07T00:00:00
[ [ "Zhang", "Zheyuan", "" ], [ "Wang", "Bin", "" ], [ "Yao", "Lanhong", "" ], [ "Demir", "Ugur", "" ], [ "Jha", "Debesh", "" ], [ "Turkbey", "Ismail Baris", "" ], [ "Gong", "Boqing", "" ], [ "Bagci", "Ulas", "" ] ]
TITLE: Domain Generalization with Adversarial Intensity Attack for Medical Image Segmentation ABSTRACT: Most statistical learning algorithms rely on an over-simplified assumption, that is, the train and test data are independent and identically distributed. In real-world scenarios, however, it is common for models to encounter data from new and different domains to which they were not exposed to during training. This is often the case in medical imaging applications due to differences in acquisition devices, imaging protocols, and patient characteristics. To address this problem, domain generalization (DG) is a promising direction as it enables models to handle data from previously unseen domains by learning domain-invariant features robust to variations across different domains. To this end, we introduce a novel DG method called Adversarial Intensity Attack (AdverIN), which leverages adversarial training to generate training data with an infinite number of styles and increase data diversity while preserving essential content information. We conduct extensive evaluation experiments on various multi-domain segmentation datasets, including 2D retinal fundus optic disc/cup and 3D prostate MRI. Our results demonstrate that AdverIN significantly improves the generalization ability of the segmentation models, achieving significant improvement on these challenging datasets. Code is available upon publication.
no_new_dataset
0.709655
2304.02724
Blake VanBerlo
Blake VanBerlo, Brian Li, Alexander Wong, Jesse Hoey, Robert Arntfield
Exploring the Utility of Self-Supervised Pretraining Strategies for the Detection of Absent Lung Sliding in M-Mode Lung Ultrasound
10 pages, 6 figures, to be published in the 2023 Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops
null
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Self-supervised pretraining has been observed to improve performance in supervised learning tasks in medical imaging. This study investigates the utility of self-supervised pretraining prior to conducting supervised fine-tuning for the downstream task of lung sliding classification in M-mode lung ultrasound images. We propose a novel pairwise relationship that couples M-mode images constructed from the same B-mode image and investigate the utility of data augmentation procedure specific to M-mode lung ultrasound. The results indicate that self-supervised pretraining yields better performance than full supervision, most notably for feature extractors not initialized with ImageNet-pretrained weights. Moreover, we observe that including a vast volume of unlabelled data results in improved performance on external validation datasets, underscoring the value of self-supervision for improving generalizability in automatic ultrasound interpretation. To the authors' best knowledge, this study is the first to characterize the influence of self-supervised pretraining for M-mode ultrasound.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 20:01:59 GMT" } ]
2023-04-07T00:00:00
[ [ "VanBerlo", "Blake", "" ], [ "Li", "Brian", "" ], [ "Wong", "Alexander", "" ], [ "Hoey", "Jesse", "" ], [ "Arntfield", "Robert", "" ] ]
TITLE: Exploring the Utility of Self-Supervised Pretraining Strategies for the Detection of Absent Lung Sliding in M-Mode Lung Ultrasound ABSTRACT: Self-supervised pretraining has been observed to improve performance in supervised learning tasks in medical imaging. This study investigates the utility of self-supervised pretraining prior to conducting supervised fine-tuning for the downstream task of lung sliding classification in M-mode lung ultrasound images. We propose a novel pairwise relationship that couples M-mode images constructed from the same B-mode image and investigate the utility of data augmentation procedure specific to M-mode lung ultrasound. The results indicate that self-supervised pretraining yields better performance than full supervision, most notably for feature extractors not initialized with ImageNet-pretrained weights. Moreover, we observe that including a vast volume of unlabelled data results in improved performance on external validation datasets, underscoring the value of self-supervision for improving generalizability in automatic ultrasound interpretation. To the authors' best knowledge, this study is the first to characterize the influence of self-supervised pretraining for M-mode ultrasound.
no_new_dataset
0.712013
2304.02729
Giulia Bernardini
Giulia Bernardini and Leo van Iersel and Esther Julien and Leen Stougie
Constructing Phylogenetic Networks via Cherry Picking and Machine Learning
42 pages, 20 figures, submitted to Algorithms for Molecular Biology (special issue of WABI 2022)
null
null
null
q-bio.PE cs.LG
http://creativecommons.org/licenses/by/4.0/
Combining a set of phylogenetic trees into a single phylogenetic network that explains all of them is a fundamental challenge in evolutionary studies. Existing methods are computationally expensive and can either handle only small numbers of phylogenetic trees or are limited to severely restricted classes of networks. In this paper, we apply the recently-introduced theoretical framework of cherry picking to design a class of efficient heuristics that are guaranteed to produce a network containing each of the input trees, for datasets consisting of binary trees. Some of the heuristics in this framework are based on the design and training of a machine learning model that captures essential information on the structure of the input trees and guides the algorithms towards better solutions. We also propose simple and fast randomised heuristics that prove to be very effective when run multiple times. Unlike the existing exact methods, our heuristics are applicable to datasets of practical size, and the experimental study we conducted on both simulated and real data shows that these solutions are qualitatively good, always within some small constant factor from the optimum. Moreover, our machine-learned heuristics are one of the first applications of machine learning to phylogenetics and show its promise.
[ { "version": "v1", "created": "Fri, 31 Mar 2023 15:04:42 GMT" } ]
2023-04-07T00:00:00
[ [ "Bernardini", "Giulia", "" ], [ "van Iersel", "Leo", "" ], [ "Julien", "Esther", "" ], [ "Stougie", "Leen", "" ] ]
TITLE: Constructing Phylogenetic Networks via Cherry Picking and Machine Learning ABSTRACT: Combining a set of phylogenetic trees into a single phylogenetic network that explains all of them is a fundamental challenge in evolutionary studies. Existing methods are computationally expensive and can either handle only small numbers of phylogenetic trees or are limited to severely restricted classes of networks. In this paper, we apply the recently-introduced theoretical framework of cherry picking to design a class of efficient heuristics that are guaranteed to produce a network containing each of the input trees, for datasets consisting of binary trees. Some of the heuristics in this framework are based on the design and training of a machine learning model that captures essential information on the structure of the input trees and guides the algorithms towards better solutions. We also propose simple and fast randomised heuristics that prove to be very effective when run multiple times. Unlike the existing exact methods, our heuristics are applicable to datasets of practical size, and the experimental study we conducted on both simulated and real data shows that these solutions are qualitatively good, always within some small constant factor from the optimum. Moreover, our machine-learned heuristics are one of the first applications of machine learning to phylogenetics and show its promise.
no_new_dataset
0.708843
2304.02738
Jonathan Francis
Jonathan Francis, Nariaki Kitamura, Felix Labelle, Xiaopeng Lu, Ingrid Navarro, Jean Oh
Core Challenges in Embodied Vision-Language Planning
Extended Abstract accepted to the 32nd International Joint Conference on Artificial Intelligence (IJCAI 2023); special journal track for authors of published JAIR 2022 and AIJ 2022 papers. 6 pages, 2 figures. arXiv admin note: substantial text overlap with arXiv:2106.13948
null
null
null
cs.RO cs.AI cs.CL cs.CV cs.HC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recent advances in the areas of Multimodal Machine Learning and Artificial Intelligence (AI) have led to the development of challenging tasks at the intersection of Computer Vision, Natural Language Processing, and Robotics. Whereas many approaches and previous survey pursuits have characterised one or two of these dimensions, there has not been a holistic analysis at the center of all three. Moreover, even when combinations of these topics are considered, more focus is placed on describing, e.g., current architectural methods, as opposed to also illustrating high-level challenges and opportunities for the field. In this survey paper, we discuss Embodied Vision-Language Planning (EVLP) tasks, a family of prominent embodied navigation and manipulation problems that jointly leverage computer vision and natural language for interaction in physical environments. We propose a taxonomy to unify these tasks and provide an in-depth analysis and comparison of the current and new algorithmic approaches, metrics, simulators, and datasets used for EVLP tasks. Finally, we present the core challenges that we believe new EVLP works should seek to address, and we advocate for task construction that enables model generalisability and furthers real-world deployment.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 20:37:13 GMT" } ]
2023-04-07T00:00:00
[ [ "Francis", "Jonathan", "" ], [ "Kitamura", "Nariaki", "" ], [ "Labelle", "Felix", "" ], [ "Lu", "Xiaopeng", "" ], [ "Navarro", "Ingrid", "" ], [ "Oh", "Jean", "" ] ]
TITLE: Core Challenges in Embodied Vision-Language Planning ABSTRACT: Recent advances in the areas of Multimodal Machine Learning and Artificial Intelligence (AI) have led to the development of challenging tasks at the intersection of Computer Vision, Natural Language Processing, and Robotics. Whereas many approaches and previous survey pursuits have characterised one or two of these dimensions, there has not been a holistic analysis at the center of all three. Moreover, even when combinations of these topics are considered, more focus is placed on describing, e.g., current architectural methods, as opposed to also illustrating high-level challenges and opportunities for the field. In this survey paper, we discuss Embodied Vision-Language Planning (EVLP) tasks, a family of prominent embodied navigation and manipulation problems that jointly leverage computer vision and natural language for interaction in physical environments. We propose a taxonomy to unify these tasks and provide an in-depth analysis and comparison of the current and new algorithmic approaches, metrics, simulators, and datasets used for EVLP tasks. Finally, we present the core challenges that we believe new EVLP works should seek to address, and we advocate for task construction that enables model generalisability and furthers real-world deployment.
no_new_dataset
0.707569
2304.02755
Justin Ruths
Joshua Ortiz, Alyssa Vellucci, Justin Koeln, Justin Ruths
Hybrid Zonotopes Exactly Represent ReLU Neural Networks
null
null
null
null
cs.LG cs.SY eess.SY
http://creativecommons.org/licenses/by/4.0/
We show that hybrid zonotopes offer an equivalent representation of feed-forward fully connected neural networks with ReLU activation functions. Our approach demonstrates that the complexity of binary variables is equal to the total number of neurons in the network and hence grows linearly in the size of the network. We demonstrate the utility of the hybrid zonotope formulation through three case studies including nonlinear function approximation, MPC closed-loop reachability and verification, and robustness of classification on the MNIST dataset.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 21:39:00 GMT" } ]
2023-04-07T00:00:00
[ [ "Ortiz", "Joshua", "" ], [ "Vellucci", "Alyssa", "" ], [ "Koeln", "Justin", "" ], [ "Ruths", "Justin", "" ] ]
TITLE: Hybrid Zonotopes Exactly Represent ReLU Neural Networks ABSTRACT: We show that hybrid zonotopes offer an equivalent representation of feed-forward fully connected neural networks with ReLU activation functions. Our approach demonstrates that the complexity of binary variables is equal to the total number of neurons in the network and hence grows linearly in the size of the network. We demonstrate the utility of the hybrid zonotope formulation through three case studies including nonlinear function approximation, MPC closed-loop reachability and verification, and robustness of classification on the MNIST dataset.
no_new_dataset
0.71039
2304.02757
Hend Al-Khalifa Prof.
Hend Al-Khalifa, Malak Mashaabi, Ghadi Al-Yahya and Raghad Alnashwan
The Saudi Privacy Policy Dataset
8 pages, 1 figure
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by-nc-sa/4.0/
This paper introduces the Saudi Privacy Policy Dataset, a diverse compilation of Arabic privacy policies from various sectors in Saudi Arabia, annotated according to the 10 principles of the Personal Data Protection Law (PDPL); the PDPL was established to be compatible with General Data Protection Regulation (GDPR); one of the most comprehensive data regulations worldwide. Data were collected from multiple sources, including the Saudi Central Bank, the Saudi Arabia National United Platform, the Council of Health Insurance, and general websites using Google and Wikipedia. The final dataset includes 1,000 websites belonging to 7 sectors, 4,638 lines of text, 775,370 tokens, and a corpus size of 8,353 KB. The annotated dataset offers significant reuse potential for assessing privacy policy compliance, benchmarking privacy practices across industries, and developing automated tools for monitoring adherence to data protection regulations. By providing a comprehensive and annotated dataset of privacy policies, this paper aims to facilitate further research and development in the areas of privacy policy analysis, natural language processing, and machine learning applications related to privacy and data protection, while also serving as an essential resource for researchers, policymakers, and industry professionals interested in understanding and promoting compliance with privacy regulations in Saudi Arabia.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 21:40:37 GMT" } ]
2023-04-07T00:00:00
[ [ "Al-Khalifa", "Hend", "" ], [ "Mashaabi", "Malak", "" ], [ "Al-Yahya", "Ghadi", "" ], [ "Alnashwan", "Raghad", "" ] ]
TITLE: The Saudi Privacy Policy Dataset ABSTRACT: This paper introduces the Saudi Privacy Policy Dataset, a diverse compilation of Arabic privacy policies from various sectors in Saudi Arabia, annotated according to the 10 principles of the Personal Data Protection Law (PDPL); the PDPL was established to be compatible with General Data Protection Regulation (GDPR); one of the most comprehensive data regulations worldwide. Data were collected from multiple sources, including the Saudi Central Bank, the Saudi Arabia National United Platform, the Council of Health Insurance, and general websites using Google and Wikipedia. The final dataset includes 1,000 websites belonging to 7 sectors, 4,638 lines of text, 775,370 tokens, and a corpus size of 8,353 KB. The annotated dataset offers significant reuse potential for assessing privacy policy compliance, benchmarking privacy practices across industries, and developing automated tools for monitoring adherence to data protection regulations. By providing a comprehensive and annotated dataset of privacy policies, this paper aims to facilitate further research and development in the areas of privacy policy analysis, natural language processing, and machine learning applications related to privacy and data protection, while also serving as an essential resource for researchers, policymakers, and industry professionals interested in understanding and promoting compliance with privacy regulations in Saudi Arabia.
new_dataset
0.715057
2304.02767
Satish Kumar
Satish Kumar, Ivan Arevalo, ASM Iftekhar, B S Manjunath
MethaneMapper: Spectral Absorption aware Hyperspectral Transformer for Methane Detection
16 pages, 9 figures, 3 tables, Accepted to Computer Vision and Pattern Recognition (CVPR 2023) conference. CVPR 2023 Highlights paper
null
null
null
eess.IV cs.CV cs.LG stat.ML
http://creativecommons.org/licenses/by/4.0/
Methane (CH$_4$) is the chief contributor to global climate change. Recent Airborne Visible-Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) has been very useful in quantitative mapping of methane emissions. Existing methods for analyzing this data are sensitive to local terrain conditions, often require manual inspection from domain experts, prone to significant error and hence are not scalable. To address these challenges, we propose a novel end-to-end spectral absorption wavelength aware transformer network, MethaneMapper, to detect and quantify the emissions. MethaneMapper introduces two novel modules that help to locate the most relevant methane plume regions in the spectral domain and uses them to localize these accurately. Thorough evaluation shows that MethaneMapper achieves 0.63 mAP in detection and reduces the model size (by 5x) compared to the current state of the art. In addition, we also introduce a large-scale dataset of methane plume segmentation mask for over 1200 AVIRIS-NG flight lines from 2015-2022. It contains over 4000 methane plume sites. Our dataset will provide researchers the opportunity to develop and advance new methods for tackling this challenging green-house gas detection problem with significant broader social impact. Dataset and source code are public
[ { "version": "v1", "created": "Wed, 5 Apr 2023 22:15:18 GMT" } ]
2023-04-07T00:00:00
[ [ "Kumar", "Satish", "" ], [ "Arevalo", "Ivan", "" ], [ "Iftekhar", "ASM", "" ], [ "Manjunath", "B S", "" ] ]
TITLE: MethaneMapper: Spectral Absorption aware Hyperspectral Transformer for Methane Detection ABSTRACT: Methane (CH$_4$) is the chief contributor to global climate change. Recent Airborne Visible-Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) has been very useful in quantitative mapping of methane emissions. Existing methods for analyzing this data are sensitive to local terrain conditions, often require manual inspection from domain experts, prone to significant error and hence are not scalable. To address these challenges, we propose a novel end-to-end spectral absorption wavelength aware transformer network, MethaneMapper, to detect and quantify the emissions. MethaneMapper introduces two novel modules that help to locate the most relevant methane plume regions in the spectral domain and uses them to localize these accurately. Thorough evaluation shows that MethaneMapper achieves 0.63 mAP in detection and reduces the model size (by 5x) compared to the current state of the art. In addition, we also introduce a large-scale dataset of methane plume segmentation mask for over 1200 AVIRIS-NG flight lines from 2015-2022. It contains over 4000 methane plume sites. Our dataset will provide researchers the opportunity to develop and advance new methods for tackling this challenging green-house gas detection problem with significant broader social impact. Dataset and source code are public
new_dataset
0.713669
2304.02768
Vitor Batista
Vitor Alcantara Batista, Alexandre Gon\c{c}alves Evsukoff
Application of Transformers based methods in Electronic Medical Records: A Systematic Literature Review
null
null
null
null
cs.CL cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
The combined growth of available data and their unstructured nature has received increased interest in natural language processing (NLP) techniques to make value of these data assets since this format is not suitable for statistical analysis. This work presents a systematic literature review of state-of-the-art advances using transformer-based methods on electronic medical records (EMRs) in different NLP tasks. To the best of our knowledge, this work is unique in providing a comprehensive review of research on transformer-based methods for NLP applied to the EMR field. In the initial query, 99 articles were selected from three public databases and filtered into 65 articles for detailed analysis. The papers were analyzed with respect to the business problem, NLP task, models and techniques, availability of datasets, reproducibility of modeling, language, and exchange format. The paper presents some limitations of current research and some recommendations for further research.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 22:19:42 GMT" } ]
2023-04-07T00:00:00
[ [ "Batista", "Vitor Alcantara", "" ], [ "Evsukoff", "Alexandre Gonçalves", "" ] ]
TITLE: Application of Transformers based methods in Electronic Medical Records: A Systematic Literature Review ABSTRACT: The combined growth of available data and their unstructured nature has received increased interest in natural language processing (NLP) techniques to make value of these data assets since this format is not suitable for statistical analysis. This work presents a systematic literature review of state-of-the-art advances using transformer-based methods on electronic medical records (EMRs) in different NLP tasks. To the best of our knowledge, this work is unique in providing a comprehensive review of research on transformer-based methods for NLP applied to the EMR field. In the initial query, 99 articles were selected from three public databases and filtered into 65 articles for detailed analysis. The papers were analyzed with respect to the business problem, NLP task, models and techniques, availability of datasets, reproducibility of modeling, language, and exchange format. The paper presents some limitations of current research and some recommendations for further research.
no_new_dataset
0.710038
2304.02769
Viswanath Chadalapaka
Viswanath Chadalapaka, Derek Nguyen, JoonWon Choi, Shaunak Joshi, Mohammad Rostami
Low-Shot Learning for Fictional Claim Verification
6 pages
null
null
null
cs.AI
http://creativecommons.org/licenses/by/4.0/
In this paper, we study the problem of claim verification in the context of claims about fictional stories in a low-shot learning setting. To this end, we generate two synthetic datasets and then develop an end-to-end pipeline and model that is tested on both benchmarks. To test the efficacy of our pipeline and the difficulty of benchmarks, we compare our models' results against human and random assignment results. Our code is available at https://github.com/Derposoft/plot_hole_detection.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 22:20:40 GMT" } ]
2023-04-07T00:00:00
[ [ "Chadalapaka", "Viswanath", "" ], [ "Nguyen", "Derek", "" ], [ "Choi", "JoonWon", "" ], [ "Joshi", "Shaunak", "" ], [ "Rostami", "Mohammad", "" ] ]
TITLE: Low-Shot Learning for Fictional Claim Verification ABSTRACT: In this paper, we study the problem of claim verification in the context of claims about fictional stories in a low-shot learning setting. To this end, we generate two synthetic datasets and then develop an end-to-end pipeline and model that is tested on both benchmarks. To test the efficacy of our pipeline and the difficulty of benchmarks, we compare our models' results against human and random assignment results. Our code is available at https://github.com/Derposoft/plot_hole_detection.
no_new_dataset
0.712063
2304.02777
Xiaoqian Shen
Xiaoqian Shen, Xiang Li, Mohamed Elhoseiny
MoStGAN-V: Video Generation with Temporal Motion Styles
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Video generation remains a challenging task due to spatiotemporal complexity and the requirement of synthesizing diverse motions with temporal consistency. Previous works attempt to generate videos in arbitrary lengths either in an autoregressive manner or regarding time as a continuous signal. However, they struggle to synthesize detailed and diverse motions with temporal coherence and tend to generate repetitive scenes after a few time steps. In this work, we argue that a single time-agnostic latent vector of style-based generator is insufficient to model various and temporally-consistent motions. Hence, we introduce additional time-dependent motion styles to model diverse motion patterns. In addition, a Motion Style Attention modulation mechanism, dubbed as MoStAtt, is proposed to augment frames with vivid dynamics for each specific scale (i.e., layer), which assigns attention score for each motion style w.r.t deconvolution filter weights in the target synthesis layer and softly attends different motion styles for weight modulation. Experimental results show our model achieves state-of-the-art performance on four unconditional $256^2$ video synthesis benchmarks trained with only 3 frames per clip and produces better qualitative results with respect to dynamic motions. Code and videos have been made available at https://github.com/xiaoqian-shen/MoStGAN-V.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 22:47:12 GMT" } ]
2023-04-07T00:00:00
[ [ "Shen", "Xiaoqian", "" ], [ "Li", "Xiang", "" ], [ "Elhoseiny", "Mohamed", "" ] ]
TITLE: MoStGAN-V: Video Generation with Temporal Motion Styles ABSTRACT: Video generation remains a challenging task due to spatiotemporal complexity and the requirement of synthesizing diverse motions with temporal consistency. Previous works attempt to generate videos in arbitrary lengths either in an autoregressive manner or regarding time as a continuous signal. However, they struggle to synthesize detailed and diverse motions with temporal coherence and tend to generate repetitive scenes after a few time steps. In this work, we argue that a single time-agnostic latent vector of style-based generator is insufficient to model various and temporally-consistent motions. Hence, we introduce additional time-dependent motion styles to model diverse motion patterns. In addition, a Motion Style Attention modulation mechanism, dubbed as MoStAtt, is proposed to augment frames with vivid dynamics for each specific scale (i.e., layer), which assigns attention score for each motion style w.r.t deconvolution filter weights in the target synthesis layer and softly attends different motion styles for weight modulation. Experimental results show our model achieves state-of-the-art performance on four unconditional $256^2$ video synthesis benchmarks trained with only 3 frames per clip and produces better qualitative results with respect to dynamic motions. Code and videos have been made available at https://github.com/xiaoqian-shen/MoStGAN-V.
no_new_dataset
0.710069
2304.02782
Min Chen
Min Chen and Zhikun Zhang and Tianhao Wang and Michael Backes and Yang Zhang
FACE-AUDITOR: Data Auditing in Facial Recognition Systems
To appear in the 32nd USENIX Security Symposium, August 2023, Anaheim, CA, USA
null
null
null
cs.CR cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Few-shot-based facial recognition systems have gained increasing attention due to their scalability and ability to work with a few face images during the model deployment phase. However, the power of facial recognition systems enables entities with moderate resources to canvas the Internet and build well-performed facial recognition models without people's awareness and consent. To prevent the face images from being misused, one straightforward approach is to modify the raw face images before sharing them, which inevitably destroys the semantic information, increases the difficulty of retroactivity, and is still prone to adaptive attacks. Therefore, an auditing method that does not interfere with the facial recognition model's utility and cannot be quickly bypassed is urgently needed. In this paper, we formulate the auditing process as a user-level membership inference problem and propose a complete toolkit FACE-AUDITOR that can carefully choose the probing set to query the few-shot-based facial recognition model and determine whether any of a user's face images is used in training the model. We further propose to use the similarity scores between the original face images as reference information to improve the auditing performance. Extensive experiments on multiple real-world face image datasets show that FACE-AUDITOR can achieve auditing accuracy of up to $99\%$. Finally, we show that FACE-AUDITOR is robust in the presence of several perturbation mechanisms to the training images or the target models. The source code of our experiments can be found at \url{https://github.com/MinChen00/Face-Auditor}.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 23:03:54 GMT" } ]
2023-04-07T00:00:00
[ [ "Chen", "Min", "" ], [ "Zhang", "Zhikun", "" ], [ "Wang", "Tianhao", "" ], [ "Backes", "Michael", "" ], [ "Zhang", "Yang", "" ] ]
TITLE: FACE-AUDITOR: Data Auditing in Facial Recognition Systems ABSTRACT: Few-shot-based facial recognition systems have gained increasing attention due to their scalability and ability to work with a few face images during the model deployment phase. However, the power of facial recognition systems enables entities with moderate resources to canvas the Internet and build well-performed facial recognition models without people's awareness and consent. To prevent the face images from being misused, one straightforward approach is to modify the raw face images before sharing them, which inevitably destroys the semantic information, increases the difficulty of retroactivity, and is still prone to adaptive attacks. Therefore, an auditing method that does not interfere with the facial recognition model's utility and cannot be quickly bypassed is urgently needed. In this paper, we formulate the auditing process as a user-level membership inference problem and propose a complete toolkit FACE-AUDITOR that can carefully choose the probing set to query the few-shot-based facial recognition model and determine whether any of a user's face images is used in training the model. We further propose to use the similarity scores between the original face images as reference information to improve the auditing performance. Extensive experiments on multiple real-world face image datasets show that FACE-AUDITOR can achieve auditing accuracy of up to $99\%$. Finally, we show that FACE-AUDITOR is robust in the presence of several perturbation mechanisms to the training images or the target models. The source code of our experiments can be found at \url{https://github.com/MinChen00/Face-Auditor}.
no_new_dataset
0.709202
2304.02797
Vitor Guizilini
Vitor Guizilini, Igor Vasiljevic, Jiading Fang, Rares Ambrus, Sergey Zakharov, Vincent Sitzmann, Adrien Gaidon
DeLiRa: Self-Supervised Depth, Light, and Radiance Fields
Project page: https://sites.google.com/view/tri-delira
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Differentiable volumetric rendering is a powerful paradigm for 3D reconstruction and novel view synthesis. However, standard volume rendering approaches struggle with degenerate geometries in the case of limited viewpoint diversity, a common scenario in robotics applications. In this work, we propose to use the multi-view photometric objective from the self-supervised depth estimation literature as a geometric regularizer for volumetric rendering, significantly improving novel view synthesis without requiring additional information. Building upon this insight, we explore the explicit modeling of scene geometry using a generalist Transformer, jointly learning a radiance field as well as depth and light fields with a set of shared latent codes. We demonstrate that sharing geometric information across tasks is mutually beneficial, leading to improvements over single-task learning without an increase in network complexity. Our DeLiRa architecture achieves state-of-the-art results on the ScanNet benchmark, enabling high quality volumetric rendering as well as real-time novel view and depth synthesis in the limited viewpoint diversity setting.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 00:16:25 GMT" } ]
2023-04-07T00:00:00
[ [ "Guizilini", "Vitor", "" ], [ "Vasiljevic", "Igor", "" ], [ "Fang", "Jiading", "" ], [ "Ambrus", "Rares", "" ], [ "Zakharov", "Sergey", "" ], [ "Sitzmann", "Vincent", "" ], [ "Gaidon", "Adrien", "" ] ]
TITLE: DeLiRa: Self-Supervised Depth, Light, and Radiance Fields ABSTRACT: Differentiable volumetric rendering is a powerful paradigm for 3D reconstruction and novel view synthesis. However, standard volume rendering approaches struggle with degenerate geometries in the case of limited viewpoint diversity, a common scenario in robotics applications. In this work, we propose to use the multi-view photometric objective from the self-supervised depth estimation literature as a geometric regularizer for volumetric rendering, significantly improving novel view synthesis without requiring additional information. Building upon this insight, we explore the explicit modeling of scene geometry using a generalist Transformer, jointly learning a radiance field as well as depth and light fields with a set of shared latent codes. We demonstrate that sharing geometric information across tasks is mutually beneficial, leading to improvements over single-task learning without an increase in network complexity. Our DeLiRa architecture achieves state-of-the-art results on the ScanNet benchmark, enabling high quality volumetric rendering as well as real-time novel view and depth synthesis in the limited viewpoint diversity setting.
no_new_dataset
0.711387
2304.02801
Fangping Xie
Fangping Xie, Pierre Le Meur, Charith Fernando
End-to-end Manipulator Calligraphy Planning via Variational Imitation Learning
5 pages, 4 figures
null
null
null
cs.RO cs.AI
http://creativecommons.org/licenses/by/4.0/
Planning from demonstrations has shown promising results with the advances of deep neural networks. One of the most popular real-world applications is automated handwriting using a robotic manipulator. Classically it is simplified as a two-dimension problem. This representation is suitable for elementary drawings, but it is not sufficient for Japanese calligraphy or complex work of art where the orientation of a pen is part of the user expression. In this study, we focus on automated planning of Japanese calligraphy using a three-dimension representation of the trajectory as well as the rotation of the pen tip, and propose a novel deep imitation learning neural network that learns from expert demonstrations through a combination of images and pose data. The network consists of a combination of variational auto-encoder, bi-directional LSTM, and Multi-Layer Perceptron (MLP). Experiments are conducted in a progressive way, and results demonstrate that the proposed approach is successful in completion of tasks for real-world robots, overcoming the distribution shift problem in imitation learning. The source code and dataset will be public.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 00:34:15 GMT" } ]
2023-04-07T00:00:00
[ [ "Xie", "Fangping", "" ], [ "Meur", "Pierre Le", "" ], [ "Fernando", "Charith", "" ] ]
TITLE: End-to-end Manipulator Calligraphy Planning via Variational Imitation Learning ABSTRACT: Planning from demonstrations has shown promising results with the advances of deep neural networks. One of the most popular real-world applications is automated handwriting using a robotic manipulator. Classically it is simplified as a two-dimension problem. This representation is suitable for elementary drawings, but it is not sufficient for Japanese calligraphy or complex work of art where the orientation of a pen is part of the user expression. In this study, we focus on automated planning of Japanese calligraphy using a three-dimension representation of the trajectory as well as the rotation of the pen tip, and propose a novel deep imitation learning neural network that learns from expert demonstrations through a combination of images and pose data. The network consists of a combination of variational auto-encoder, bi-directional LSTM, and Multi-Layer Perceptron (MLP). Experiments are conducted in a progressive way, and results demonstrate that the proposed approach is successful in completion of tasks for real-world robots, overcoming the distribution shift problem in imitation learning. The source code and dataset will be public.
no_new_dataset
0.709447
2304.02810
Kurt Thomas
Kurt Thomas, Sarah Meiklejohn, Michael A. Specter, Xiang Wang, Xavier Llor\`a, Stephan Somogyi, and David Kleidermacher
Robust, privacy-preserving, transparent, and auditable on-device blocklisting
null
null
null
null
cs.CR
http://creativecommons.org/licenses/by/4.0/
With the accelerated adoption of end-to-end encryption, there is an opportunity to re-architect security and anti-abuse primitives in a manner that preserves new privacy expectations. In this paper, we consider two novel protocols for on-device blocklisting that allow a client to determine whether an object (e.g., URL, document, image, etc.) is harmful based on threat information possessed by a so-called remote enforcer in a way that is both privacy-preserving and trustworthy. Our protocols leverage a unique combination of private set intersection to promote privacy, cryptographic hashes to ensure resilience to false positives, cryptographic signatures to improve transparency, and Merkle inclusion proofs to ensure consistency and auditability. We benchmark our protocols -- one that is time-efficient, and the other space-efficient -- to demonstrate their practical use for applications such as email, messaging, storage, and other applications. We also highlight remaining challenges, such as privacy and censorship tensions that exist with logging or reporting. We consider our work to be a critical first step towards enabling complex, multi-stakeholder discussions on how best to provide on-device protections.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 01:19:47 GMT" } ]
2023-04-07T00:00:00
[ [ "Thomas", "Kurt", "" ], [ "Meiklejohn", "Sarah", "" ], [ "Specter", "Michael A.", "" ], [ "Wang", "Xiang", "" ], [ "Llorà", "Xavier", "" ], [ "Somogyi", "Stephan", "" ], [ "Kleidermacher", "David", "" ] ]
TITLE: Robust, privacy-preserving, transparent, and auditable on-device blocklisting ABSTRACT: With the accelerated adoption of end-to-end encryption, there is an opportunity to re-architect security and anti-abuse primitives in a manner that preserves new privacy expectations. In this paper, we consider two novel protocols for on-device blocklisting that allow a client to determine whether an object (e.g., URL, document, image, etc.) is harmful based on threat information possessed by a so-called remote enforcer in a way that is both privacy-preserving and trustworthy. Our protocols leverage a unique combination of private set intersection to promote privacy, cryptographic hashes to ensure resilience to false positives, cryptographic signatures to improve transparency, and Merkle inclusion proofs to ensure consistency and auditability. We benchmark our protocols -- one that is time-efficient, and the other space-efficient -- to demonstrate their practical use for applications such as email, messaging, storage, and other applications. We also highlight remaining challenges, such as privacy and censorship tensions that exist with logging or reporting. We consider our work to be a critical first step towards enabling complex, multi-stakeholder discussions on how best to provide on-device protections.
no_new_dataset
0.706646
2304.02812
Katherine Stasaski
Katherine Stasaski and Marti A. Hearst
Pragmatically Appropriate Diversity for Dialogue Evaluation
null
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
Linguistic pragmatics state that a conversation's underlying speech acts can constrain the type of response which is appropriate at each turn in the conversation. When generating dialogue responses, neural dialogue agents struggle to produce diverse responses. Currently, dialogue diversity is assessed using automatic metrics, but the underlying speech acts do not inform these metrics. To remedy this, we propose the notion of Pragmatically Appropriate Diversity, defined as the extent to which a conversation creates and constrains the creation of multiple diverse responses. Using a human-created multi-response dataset, we find significant support for the hypothesis that speech acts provide a signal for the diversity of the set of next responses. Building on this result, we propose a new human evaluation task where creative writers predict the extent to which conversations inspire the creation of multiple diverse responses. Our studies find that writers' judgments align with the Pragmatically Appropriate Diversity of conversations. Our work suggests that expectations for diversity metric scores should vary depending on the speech act.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 01:24:18 GMT" } ]
2023-04-07T00:00:00
[ [ "Stasaski", "Katherine", "" ], [ "Hearst", "Marti A.", "" ] ]
TITLE: Pragmatically Appropriate Diversity for Dialogue Evaluation ABSTRACT: Linguistic pragmatics state that a conversation's underlying speech acts can constrain the type of response which is appropriate at each turn in the conversation. When generating dialogue responses, neural dialogue agents struggle to produce diverse responses. Currently, dialogue diversity is assessed using automatic metrics, but the underlying speech acts do not inform these metrics. To remedy this, we propose the notion of Pragmatically Appropriate Diversity, defined as the extent to which a conversation creates and constrains the creation of multiple diverse responses. Using a human-created multi-response dataset, we find significant support for the hypothesis that speech acts provide a signal for the diversity of the set of next responses. Building on this result, we propose a new human evaluation task where creative writers predict the extent to which conversations inspire the creation of multiple diverse responses. Our studies find that writers' judgments align with the Pragmatically Appropriate Diversity of conversations. Our work suggests that expectations for diversity metric scores should vary depending on the speech act.
new_dataset
0.623635
2304.02828
Noa Garcia
Noa Garcia, Yusuke Hirota, Yankun Wu, Yuta Nakashima
Uncurated Image-Text Datasets: Shedding Light on Demographic Bias
CVPR 2023
null
null
null
cs.CV cs.CY
http://creativecommons.org/licenses/by/4.0/
The increasing tendency to collect large and uncurated datasets to train vision-and-language models has raised concerns about fair representations. It is known that even small but manually annotated datasets, such as MSCOCO, are affected by societal bias. This problem, far from being solved, may be getting worse with data crawled from the Internet without much control. In addition, the lack of tools to analyze societal bias in big collections of images makes addressing the problem extremely challenging. Our first contribution is to annotate part of the Google Conceptual Captions dataset, widely used for training vision-and-language models, with four demographic and two contextual attributes. Our second contribution is to conduct a comprehensive analysis of the annotations, focusing on how different demographic groups are represented. Our last contribution lies in evaluating three prevailing vision-and-language tasks: image captioning, text-image CLIP embeddings, and text-to-image generation, showing that societal bias is a persistent problem in all of them.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 02:33:51 GMT" } ]
2023-04-07T00:00:00
[ [ "Garcia", "Noa", "" ], [ "Hirota", "Yusuke", "" ], [ "Wu", "Yankun", "" ], [ "Nakashima", "Yuta", "" ] ]
TITLE: Uncurated Image-Text Datasets: Shedding Light on Demographic Bias ABSTRACT: The increasing tendency to collect large and uncurated datasets to train vision-and-language models has raised concerns about fair representations. It is known that even small but manually annotated datasets, such as MSCOCO, are affected by societal bias. This problem, far from being solved, may be getting worse with data crawled from the Internet without much control. In addition, the lack of tools to analyze societal bias in big collections of images makes addressing the problem extremely challenging. Our first contribution is to annotate part of the Google Conceptual Captions dataset, widely used for training vision-and-language models, with four demographic and two contextual attributes. Our second contribution is to conduct a comprehensive analysis of the annotations, focusing on how different demographic groups are represented. Our last contribution lies in evaluating three prevailing vision-and-language tasks: image captioning, text-image CLIP embeddings, and text-to-image generation, showing that societal bias is a persistent problem in all of them.
no_new_dataset
0.708862
2304.02835
Jiancan Wu
Jiancan Wu, Yi Yang, Yuchun Qian, Yongduo Sui, Xiang Wang, Xiangnan He
GIF: A General Graph Unlearning Strategy via Influence Function
Accepted by WWW 2023
null
10.1145/3543507.3583521
null
cs.LG cs.SI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
With the greater emphasis on privacy and security in our society, the problem of graph unlearning -- revoking the influence of specific data on the trained GNN model, is drawing increasing attention. However, ranging from machine unlearning to recently emerged graph unlearning methods, existing efforts either resort to retraining paradigm, or perform approximate erasure that fails to consider the inter-dependency between connected neighbors or imposes constraints on GNN structure, therefore hard to achieve satisfying performance-complexity trade-offs. In this work, we explore the influence function tailored for graph unlearning, so as to improve the unlearning efficacy and efficiency for graph unlearning. We first present a unified problem formulation of diverse graph unlearning tasks \wrt node, edge, and feature. Then, we recognize the crux to the inability of traditional influence function for graph unlearning, and devise Graph Influence Function (GIF), a model-agnostic unlearning method that can efficiently and accurately estimate parameter changes in response to a $\epsilon$-mass perturbation in deleted data. The idea is to supplement the objective of the traditional influence function with an additional loss term of the influenced neighbors due to the structural dependency. Further deductions on the closed-form solution of parameter changes provide a better understanding of the unlearning mechanism. We conduct extensive experiments on four representative GNN models and three benchmark datasets to justify the superiority of GIF for diverse graph unlearning tasks in terms of unlearning efficacy, model utility, and unlearning efficiency. Our implementations are available at \url{https://github.com/wujcan/GIF-torch/}.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 03:02:54 GMT" } ]
2023-04-07T00:00:00
[ [ "Wu", "Jiancan", "" ], [ "Yang", "Yi", "" ], [ "Qian", "Yuchun", "" ], [ "Sui", "Yongduo", "" ], [ "Wang", "Xiang", "" ], [ "He", "Xiangnan", "" ] ]
TITLE: GIF: A General Graph Unlearning Strategy via Influence Function ABSTRACT: With the greater emphasis on privacy and security in our society, the problem of graph unlearning -- revoking the influence of specific data on the trained GNN model, is drawing increasing attention. However, ranging from machine unlearning to recently emerged graph unlearning methods, existing efforts either resort to retraining paradigm, or perform approximate erasure that fails to consider the inter-dependency between connected neighbors or imposes constraints on GNN structure, therefore hard to achieve satisfying performance-complexity trade-offs. In this work, we explore the influence function tailored for graph unlearning, so as to improve the unlearning efficacy and efficiency for graph unlearning. We first present a unified problem formulation of diverse graph unlearning tasks \wrt node, edge, and feature. Then, we recognize the crux to the inability of traditional influence function for graph unlearning, and devise Graph Influence Function (GIF), a model-agnostic unlearning method that can efficiently and accurately estimate parameter changes in response to a $\epsilon$-mass perturbation in deleted data. The idea is to supplement the objective of the traditional influence function with an additional loss term of the influenced neighbors due to the structural dependency. Further deductions on the closed-form solution of parameter changes provide a better understanding of the unlearning mechanism. We conduct extensive experiments on four representative GNN models and three benchmark datasets to justify the superiority of GIF for diverse graph unlearning tasks in terms of unlearning efficacy, model utility, and unlearning efficiency. Our implementations are available at \url{https://github.com/wujcan/GIF-torch/}.
no_new_dataset
0.708042
2304.02838
Xuezhi Wen
Nan Wang, Xuezhi Wen, Dalin Zhang, Xibin Zhao, Jiahui Ma, Mengxia Luo, Sen Nie, Shi Wu, Jiqiang Liu
TBDetector:Transformer-Based Detector for Advanced Persistent Threats with Provenance Graph
10 pages, 7 figures
null
null
null
cs.CR cs.AI cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
APT detection is difficult to detect due to the long-term latency, covert and slow multistage attack patterns of Advanced Persistent Threat (APT). To tackle these issues, we propose TBDetector, a transformer-based advanced persistent threat detection method for APT attack detection. Considering that provenance graphs provide rich historical information and have the powerful attacks historic correlation ability to identify anomalous activities, TBDetector employs provenance analysis for APT detection, which summarizes long-running system execution with space efficiency and utilizes transformer with self-attention based encoder-decoder to extract long-term contextual features of system states to detect slow-acting attacks. Furthermore, we further introduce anomaly scores to investigate the anomaly of different system states, where each state is calculated with an anomaly score corresponding to its similarity score and isolation score. To evaluate the effectiveness of the proposed method, we have conducted experiments on five public datasets, i.e., streamspot, cadets, shellshock, clearscope, and wget_baseline. Experimental results and comparisons with state-of-the-art methods have exhibited better performance of our proposed method.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 03:08:09 GMT" } ]
2023-04-07T00:00:00
[ [ "Wang", "Nan", "" ], [ "Wen", "Xuezhi", "" ], [ "Zhang", "Dalin", "" ], [ "Zhao", "Xibin", "" ], [ "Ma", "Jiahui", "" ], [ "Luo", "Mengxia", "" ], [ "Nie", "Sen", "" ], [ "Wu", "Shi", "" ], [ "Liu", "Jiqiang", "" ] ]
TITLE: TBDetector:Transformer-Based Detector for Advanced Persistent Threats with Provenance Graph ABSTRACT: APT detection is difficult to detect due to the long-term latency, covert and slow multistage attack patterns of Advanced Persistent Threat (APT). To tackle these issues, we propose TBDetector, a transformer-based advanced persistent threat detection method for APT attack detection. Considering that provenance graphs provide rich historical information and have the powerful attacks historic correlation ability to identify anomalous activities, TBDetector employs provenance analysis for APT detection, which summarizes long-running system execution with space efficiency and utilizes transformer with self-attention based encoder-decoder to extract long-term contextual features of system states to detect slow-acting attacks. Furthermore, we further introduce anomaly scores to investigate the anomaly of different system states, where each state is calculated with an anomaly score corresponding to its similarity score and isolation score. To evaluate the effectiveness of the proposed method, we have conducted experiments on five public datasets, i.e., streamspot, cadets, shellshock, clearscope, and wget_baseline. Experimental results and comparisons with state-of-the-art methods have exhibited better performance of our proposed method.
no_new_dataset
0.709799
2304.02840
Yite Wang
Yite Wang, Dawei Li, Ruoyu Sun
NTK-SAP: Improving neural network pruning by aligning training dynamics
Accepted at ICLR 2023 (https://openreview.net/forum?id=-5EWhW_4qWP). Our code will be released at https://github.com/YiteWang/NTK-SAP
null
null
null
cs.LG cs.CV stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Pruning neural networks before training has received increasing interest due to its potential to reduce training time and memory. One popular method is to prune the connections based on a certain metric, but it is not entirely clear what metric is the best choice. Recent advances in neural tangent kernel (NTK) theory suggest that the training dynamics of large enough neural networks is closely related to the spectrum of the NTK. Motivated by this finding, we propose to prune the connections that have the least influence on the spectrum of the NTK. This method can help maintain the NTK spectrum, which may help align the training dynamics to that of its dense counterpart. However, one possible issue is that the fixed-weight-NTK corresponding to a given initial point can be very different from the NTK corresponding to later iterates during the training phase. We further propose to sample multiple realizations of random weights to estimate the NTK spectrum. Note that our approach is weight-agnostic, which is different from most existing methods that are weight-dependent. In addition, we use random inputs to compute the fixed-weight-NTK, making our method data-agnostic as well. We name our foresight pruning algorithm Neural Tangent Kernel Spectrum-Aware Pruning (NTK-SAP). Empirically, our method achieves better performance than all baselines on multiple datasets.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 03:10:03 GMT" } ]
2023-04-07T00:00:00
[ [ "Wang", "Yite", "" ], [ "Li", "Dawei", "" ], [ "Sun", "Ruoyu", "" ] ]
TITLE: NTK-SAP: Improving neural network pruning by aligning training dynamics ABSTRACT: Pruning neural networks before training has received increasing interest due to its potential to reduce training time and memory. One popular method is to prune the connections based on a certain metric, but it is not entirely clear what metric is the best choice. Recent advances in neural tangent kernel (NTK) theory suggest that the training dynamics of large enough neural networks is closely related to the spectrum of the NTK. Motivated by this finding, we propose to prune the connections that have the least influence on the spectrum of the NTK. This method can help maintain the NTK spectrum, which may help align the training dynamics to that of its dense counterpart. However, one possible issue is that the fixed-weight-NTK corresponding to a given initial point can be very different from the NTK corresponding to later iterates during the training phase. We further propose to sample multiple realizations of random weights to estimate the NTK spectrum. Note that our approach is weight-agnostic, which is different from most existing methods that are weight-dependent. In addition, we use random inputs to compute the fixed-weight-NTK, making our method data-agnostic as well. We name our foresight pruning algorithm Neural Tangent Kernel Spectrum-Aware Pruning (NTK-SAP). Empirically, our method achieves better performance than all baselines on multiple datasets.
no_new_dataset
0.710584
2304.02841
Zhijie Deng
Zhijie Deng and Yucen Luo
Learning Neural Eigenfunctions for Unsupervised Semantic Segmentation
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Unsupervised semantic segmentation is a long-standing challenge in computer vision with great significance. Spectral clustering is a theoretically grounded solution to it where the spectral embeddings for pixels are computed to construct distinct clusters. Despite recent progress in enhancing spectral clustering with powerful pre-trained models, current approaches still suffer from inefficiencies in spectral decomposition and inflexibility in applying them to the test data. This work addresses these issues by casting spectral clustering as a parametric approach that employs neural network-based eigenfunctions to produce spectral embeddings. The outputs of the neural eigenfunctions are further restricted to discrete vectors that indicate clustering assignments directly. As a result, an end-to-end NN-based paradigm of spectral clustering emerges. In practice, the neural eigenfunctions are lightweight and take the features from pre-trained models as inputs, improving training efficiency and unleashing the potential of pre-trained models for dense prediction. We conduct extensive empirical studies to validate the effectiveness of our approach and observe significant performance gains over competitive baselines on Pascal Context, Cityscapes, and ADE20K benchmarks.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 03:14:15 GMT" } ]
2023-04-07T00:00:00
[ [ "Deng", "Zhijie", "" ], [ "Luo", "Yucen", "" ] ]
TITLE: Learning Neural Eigenfunctions for Unsupervised Semantic Segmentation ABSTRACT: Unsupervised semantic segmentation is a long-standing challenge in computer vision with great significance. Spectral clustering is a theoretically grounded solution to it where the spectral embeddings for pixels are computed to construct distinct clusters. Despite recent progress in enhancing spectral clustering with powerful pre-trained models, current approaches still suffer from inefficiencies in spectral decomposition and inflexibility in applying them to the test data. This work addresses these issues by casting spectral clustering as a parametric approach that employs neural network-based eigenfunctions to produce spectral embeddings. The outputs of the neural eigenfunctions are further restricted to discrete vectors that indicate clustering assignments directly. As a result, an end-to-end NN-based paradigm of spectral clustering emerges. In practice, the neural eigenfunctions are lightweight and take the features from pre-trained models as inputs, improving training efficiency and unleashing the potential of pre-trained models for dense prediction. We conduct extensive empirical studies to validate the effectiveness of our approach and observe significant performance gains over competitive baselines on Pascal Context, Cityscapes, and ADE20K benchmarks.
no_new_dataset
0.710038
2304.02846
Shreyank N Gowda
Shreyank N Gowda
Synthetic Sample Selection for Generalized Zero-Shot Learning
Paper accepted in CVPRW 2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Generalized Zero-Shot Learning (GZSL) has emerged as a pivotal research domain in computer vision, owing to its capability to recognize objects that have not been seen during training. Despite the significant progress achieved by generative techniques in converting traditional GZSL to fully supervised learning, they tend to generate a large number of synthetic features that are often redundant, thereby increasing training time and decreasing accuracy. To address this issue, this paper proposes a novel approach for synthetic feature selection using reinforcement learning. In particular, we propose a transformer-based selector that is trained through proximal policy optimization (PPO) to select synthetic features based on the validation classification accuracy of the seen classes, which serves as a reward. The proposed method is model-agnostic and data-agnostic, making it applicable to both images and videos and versatile for diverse applications. Our experimental results demonstrate the superiority of our approach over existing feature-generating methods, yielding improved overall performance on multiple benchmarks.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 03:22:43 GMT" } ]
2023-04-07T00:00:00
[ [ "Gowda", "Shreyank N", "" ] ]
TITLE: Synthetic Sample Selection for Generalized Zero-Shot Learning ABSTRACT: Generalized Zero-Shot Learning (GZSL) has emerged as a pivotal research domain in computer vision, owing to its capability to recognize objects that have not been seen during training. Despite the significant progress achieved by generative techniques in converting traditional GZSL to fully supervised learning, they tend to generate a large number of synthetic features that are often redundant, thereby increasing training time and decreasing accuracy. To address this issue, this paper proposes a novel approach for synthetic feature selection using reinforcement learning. In particular, we propose a transformer-based selector that is trained through proximal policy optimization (PPO) to select synthetic features based on the validation classification accuracy of the seen classes, which serves as a reward. The proposed method is model-agnostic and data-agnostic, making it applicable to both images and videos and versatile for diverse applications. Our experimental results demonstrate the superiority of our approach over existing feature-generating methods, yielding improved overall performance on multiple benchmarks.
no_new_dataset
0.710823
2304.02861
Emma Pretty
Emma J. Pretty, Renan Guarese, Haytham M. Fayek, Fabio Zambetta
Replicability and Transparency for the Creation of Public Human User Video Game Datasets
Accepted for submission at the Data4XR Workshop at IEEEVR2023
null
null
null
cs.HC
http://creativecommons.org/licenses/by/4.0/
Replicability is absent in games research; a lack of transparency in protocol detail hinders scientific consensus and willingness to publish public datasets, impacting the application of these techniques in video games research. To combat this, we propose and give an example of the use of a set of experimental considerations, such as games and materials choice. This work promotes the communication of research protocols when publishing datasets, benefiting researchers when designing experiments.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 04:43:26 GMT" } ]
2023-04-07T00:00:00
[ [ "Pretty", "Emma J.", "" ], [ "Guarese", "Renan", "" ], [ "Fayek", "Haytham M.", "" ], [ "Zambetta", "Fabio", "" ] ]
TITLE: Replicability and Transparency for the Creation of Public Human User Video Game Datasets ABSTRACT: Replicability is absent in games research; a lack of transparency in protocol detail hinders scientific consensus and willingness to publish public datasets, impacting the application of these techniques in video games research. To combat this, we propose and give an example of the use of a set of experimental considerations, such as games and materials choice. This work promotes the communication of research protocols when publishing datasets, benefiting researchers when designing experiments.
no_new_dataset
0.712426
2304.02862
Sambhavi Tiwari
Sambhavi Tiwari, Manas Gogoi, Shekhar Verma, Krishna Pratap Singh
Learning to Learn with Indispensable Connections
null
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Meta-learning aims to solve unseen tasks with few labelled instances. Nevertheless, despite its effectiveness for quick learning in existing optimization-based methods, it has several flaws. Inconsequential connections are frequently seen during meta-training, which results in an over-parameterized neural network. Because of this, meta-testing observes unnecessary computations and extra memory overhead. To overcome such flaws. We propose a novel meta-learning method called Meta-LTH that includes indispensible (necessary) connections. We applied the lottery ticket hypothesis technique known as magnitude pruning to generate these crucial connections that can effectively solve few-shot learning problem. We aim to perform two things: (a) to find a sub-network capable of more adaptive meta-learning and (b) to learn new low-level features of unseen tasks and recombine those features with the already learned features during the meta-test phase. Experimental results show that our proposed Met-LTH method outperformed existing first-order MAML algorithm for three different classification datasets. Our method improves the classification accuracy by approximately 2% (20-way 1-shot task setting) for omniglot dataset.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 04:53:13 GMT" } ]
2023-04-07T00:00:00
[ [ "Tiwari", "Sambhavi", "" ], [ "Gogoi", "Manas", "" ], [ "Verma", "Shekhar", "" ], [ "Singh", "Krishna Pratap", "" ] ]
TITLE: Learning to Learn with Indispensable Connections ABSTRACT: Meta-learning aims to solve unseen tasks with few labelled instances. Nevertheless, despite its effectiveness for quick learning in existing optimization-based methods, it has several flaws. Inconsequential connections are frequently seen during meta-training, which results in an over-parameterized neural network. Because of this, meta-testing observes unnecessary computations and extra memory overhead. To overcome such flaws. We propose a novel meta-learning method called Meta-LTH that includes indispensible (necessary) connections. We applied the lottery ticket hypothesis technique known as magnitude pruning to generate these crucial connections that can effectively solve few-shot learning problem. We aim to perform two things: (a) to find a sub-network capable of more adaptive meta-learning and (b) to learn new low-level features of unseen tasks and recombine those features with the already learned features during the meta-test phase. Experimental results show that our proposed Met-LTH method outperformed existing first-order MAML algorithm for three different classification datasets. Our method improves the classification accuracy by approximately 2% (20-way 1-shot task setting) for omniglot dataset.
no_new_dataset
0.709177
2304.02870
Blandon Meng
Alexandru Rusescu, Brooke Lampe, and Weizhi Meng
Protecting User Privacy in Online Settings via Supervised Learning
null
null
null
null
cs.CR cs.LG
http://creativecommons.org/licenses/by-nc-nd/4.0/
Companies that have an online presence-in particular, companies that are exclusively digital-often subscribe to this business model: collect data from the user base, then expose the data to advertisement agencies in order to turn a profit. Such companies routinely market a service as "free", while obfuscating the fact that they tend to "charge" users in the currency of personal information rather than money. However, online companies also gather user data for more principled purposes, such as improving the user experience and aggregating statistics. The problem is the sale of user data to third parties. In this work, we design an intelligent approach to online privacy protection that leverages supervised learning. By detecting and blocking data collection that might infringe on a user's privacy, we can restore a degree of digital privacy to the user. In our evaluation, we collect a dataset of network requests and measure the performance of several classifiers that adhere to the supervised learning paradigm. The results of our evaluation demonstrate the feasibility and potential of our approach.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 05:20:16 GMT" } ]
2023-04-07T00:00:00
[ [ "Rusescu", "Alexandru", "" ], [ "Lampe", "Brooke", "" ], [ "Meng", "Weizhi", "" ] ]
TITLE: Protecting User Privacy in Online Settings via Supervised Learning ABSTRACT: Companies that have an online presence-in particular, companies that are exclusively digital-often subscribe to this business model: collect data from the user base, then expose the data to advertisement agencies in order to turn a profit. Such companies routinely market a service as "free", while obfuscating the fact that they tend to "charge" users in the currency of personal information rather than money. However, online companies also gather user data for more principled purposes, such as improving the user experience and aggregating statistics. The problem is the sale of user data to third parties. In this work, we design an intelligent approach to online privacy protection that leverages supervised learning. By detecting and blocking data collection that might infringe on a user's privacy, we can restore a degree of digital privacy to the user. In our evaluation, we collect a dataset of network requests and measure the performance of several classifiers that adhere to the supervised learning paradigm. The results of our evaluation demonstrate the feasibility and potential of our approach.
no_new_dataset
0.699421
2304.02886
David Laiymani Dr
Yakini Tchouka, Jean-Fran\c{c}ois Couchot, David Laiymani, Philippe Selles, Azzedine Rahmani
Automatic ICD-10 Code Association: A Challenging Task on French Clinical Texts
null
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by/4.0/
Automatically associating ICD codes with electronic health data is a well-known NLP task in medical research. NLP has evolved significantly in recent years with the emergence of pre-trained language models based on Transformers architecture, mainly in the English language. This paper adapts these models to automatically associate the ICD codes. Several neural network architectures have been experimented with to address the challenges of dealing with a large set of both input tokens and labels to be guessed. In this paper, we propose a model that combines the latest advances in NLP and multi-label classification for ICD-10 code association. Fair experiments on a Clinical dataset in the French language show that our approach increases the $F_1$-score metric by more than 55\% compared to state-of-the-art results.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 06:31:54 GMT" } ]
2023-04-07T00:00:00
[ [ "Tchouka", "Yakini", "" ], [ "Couchot", "Jean-François", "" ], [ "Laiymani", "David", "" ], [ "Selles", "Philippe", "" ], [ "Rahmani", "Azzedine", "" ] ]
TITLE: Automatic ICD-10 Code Association: A Challenging Task on French Clinical Texts ABSTRACT: Automatically associating ICD codes with electronic health data is a well-known NLP task in medical research. NLP has evolved significantly in recent years with the emergence of pre-trained language models based on Transformers architecture, mainly in the English language. This paper adapts these models to automatically associate the ICD codes. Several neural network architectures have been experimented with to address the challenges of dealing with a large set of both input tokens and labels to be guessed. In this paper, we propose a model that combines the latest advances in NLP and multi-label classification for ICD-10 code association. Fair experiments on a Clinical dataset in the French language show that our approach increases the $F_1$-score metric by more than 55\% compared to state-of-the-art results.
no_new_dataset
0.711005
2304.02892
Chenrui Wu
Chenrui Wu, Zexi Li, Fangxin Wang, Chao Wu
Learning Cautiously in Federated Learning with Noisy and Heterogeneous Clients
Accepted by IEEE ICME 2023
null
null
null
cs.LG cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Federated learning (FL) is a distributed framework for collaboratively training with privacy guarantees. In real-world scenarios, clients may have Non-IID data (local class imbalance) with poor annotation quality (label noise). The co-existence of label noise and class imbalance in FL's small local datasets renders conventional FL methods and noisy-label learning methods both ineffective. To address the challenges, we propose FedCNI without using an additional clean proxy dataset. It includes a noise-resilient local solver and a robust global aggregator. For the local solver, we design a more robust prototypical noise detector to distinguish noisy samples. Further to reduce the negative impact brought by the noisy samples, we devise a curriculum pseudo labeling method and a denoise Mixup training strategy. For the global aggregator, we propose a switching re-weighted aggregation method tailored to different learning periods. Extensive experiments demonstrate our method can substantially outperform state-of-the-art solutions in mix-heterogeneous FL environments.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 06:47:14 GMT" } ]
2023-04-07T00:00:00
[ [ "Wu", "Chenrui", "" ], [ "Li", "Zexi", "" ], [ "Wang", "Fangxin", "" ], [ "Wu", "Chao", "" ] ]
TITLE: Learning Cautiously in Federated Learning with Noisy and Heterogeneous Clients ABSTRACT: Federated learning (FL) is a distributed framework for collaboratively training with privacy guarantees. In real-world scenarios, clients may have Non-IID data (local class imbalance) with poor annotation quality (label noise). The co-existence of label noise and class imbalance in FL's small local datasets renders conventional FL methods and noisy-label learning methods both ineffective. To address the challenges, we propose FedCNI without using an additional clean proxy dataset. It includes a noise-resilient local solver and a robust global aggregator. For the local solver, we design a more robust prototypical noise detector to distinguish noisy samples. Further to reduce the negative impact brought by the noisy samples, we devise a curriculum pseudo labeling method and a denoise Mixup training strategy. For the global aggregator, we propose a switching re-weighted aggregation method tailored to different learning periods. Extensive experiments demonstrate our method can substantially outperform state-of-the-art solutions in mix-heterogeneous FL environments.
no_new_dataset
0.70866
2304.02897
Yiling Zeng
Yiling Zeng, Chunyao Song, Yuhan Li, Tingjian Ge
LSketch: A Label-Enabled Graph Stream Sketch Toward Time-Sensitive Queries
null
null
null
null
cs.DB cs.DS
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Graph streams represent data interactions in real applications. The mining of graph streams plays an important role in network security, social network analysis, and traffic control, among others. However, the sheer volume and high dynamics cause great challenges for efficient storage and subsequent query analysis on them. Current studies apply sketches to summarize graph streams. We propose LSketch that works for heterogeneous graph streams, which effectively preserves the label information carried by the streams in real scenes, thereby enriching the expressive ability of sketches. In addition, as graph streams continue to evolve over time, edges too old may lose their practical significance. Therefore, we introduce the sliding window model into LSketch to eliminate the expired edges automatically. LSketch uses sub-linear storage space and can support structure based queries and time-sensitive queries with high accuracy. We perform extensive experiments over four real datasets, demonstrating the superiority of the proposed method over state-of-the-art methods, in aspects of query accuracy and time efficiency.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 07:08:18 GMT" } ]
2023-04-07T00:00:00
[ [ "Zeng", "Yiling", "" ], [ "Song", "Chunyao", "" ], [ "Li", "Yuhan", "" ], [ "Ge", "Tingjian", "" ] ]
TITLE: LSketch: A Label-Enabled Graph Stream Sketch Toward Time-Sensitive Queries ABSTRACT: Graph streams represent data interactions in real applications. The mining of graph streams plays an important role in network security, social network analysis, and traffic control, among others. However, the sheer volume and high dynamics cause great challenges for efficient storage and subsequent query analysis on them. Current studies apply sketches to summarize graph streams. We propose LSketch that works for heterogeneous graph streams, which effectively preserves the label information carried by the streams in real scenes, thereby enriching the expressive ability of sketches. In addition, as graph streams continue to evolve over time, edges too old may lose their practical significance. Therefore, we introduce the sliding window model into LSketch to eliminate the expired edges automatically. LSketch uses sub-linear storage space and can support structure based queries and time-sensitive queries with high accuracy. We perform extensive experiments over four real datasets, demonstrating the superiority of the proposed method over state-of-the-art methods, in aspects of query accuracy and time efficiency.
no_new_dataset
0.70866
2304.02899
Lan Truong
Lan V. Truong
Variable-Complexity Weighted-Tempered Gibbs Samplers for Bayesian Variable Selection
29 pages, 12 figures
null
null
null
stat.ML cs.LG stat.CO stat.ME
http://creativecommons.org/licenses/by/4.0/
Subset weighted-Tempered Gibbs Sampler (wTGS) has been recently introduced by Jankowiak to reduce the computation complexity per MCMC iteration in high-dimensional applications where the exact calculation of the posterior inclusion probabilities (PIP) is not essential. However, the Rao-Backwellized estimator associated with this sampler has a high variance as the ratio between the signal dimension and the number of conditional PIP estimations is large. In this paper, we design a new subset weighted-Tempered Gibbs Sampler (wTGS) where the expected number of computations of conditional PIPs per MCMC iteration can be much smaller than the signal dimension. Different from the subset wTGS and wTGS, our sampler has a variable complexity per MCMC iteration. We provide an upper bound on the variance of an associated Rao-Blackwellized estimator for this sampler at a finite number of iterations, $T$, and show that the variance is $O\big(\big(\frac{P}{S}\big)^2 \frac{\log T}{T}\big)$ for a given dataset where $S$ is the expected number of conditional PIP computations per MCMC iteration. Experiments show that our Rao-Blackwellized estimator can have a smaller variance than its counterpart associated with the subset wTGS.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 07:14:12 GMT" } ]
2023-04-07T00:00:00
[ [ "Truong", "Lan V.", "" ] ]
TITLE: Variable-Complexity Weighted-Tempered Gibbs Samplers for Bayesian Variable Selection ABSTRACT: Subset weighted-Tempered Gibbs Sampler (wTGS) has been recently introduced by Jankowiak to reduce the computation complexity per MCMC iteration in high-dimensional applications where the exact calculation of the posterior inclusion probabilities (PIP) is not essential. However, the Rao-Backwellized estimator associated with this sampler has a high variance as the ratio between the signal dimension and the number of conditional PIP estimations is large. In this paper, we design a new subset weighted-Tempered Gibbs Sampler (wTGS) where the expected number of computations of conditional PIPs per MCMC iteration can be much smaller than the signal dimension. Different from the subset wTGS and wTGS, our sampler has a variable complexity per MCMC iteration. We provide an upper bound on the variance of an associated Rao-Blackwellized estimator for this sampler at a finite number of iterations, $T$, and show that the variance is $O\big(\big(\frac{P}{S}\big)^2 \frac{\log T}{T}\big)$ for a given dataset where $S$ is the expected number of conditional PIP computations per MCMC iteration. Experiments show that our Rao-Blackwellized estimator can have a smaller variance than its counterpart associated with the subset wTGS.
no_new_dataset
0.706634
2304.02901
Hao Zhang
Hao Zhang
SpanRE: Entities and Overlapping Relations Extraction Based on Spans and Entity Attention
null
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Extracting entities and relations is an essential task of information extraction. Triplets extracted from a sentence might overlap with each other. Previous methods either did not address the overlapping issues or solved overlapping issues partially. To tackle triplet overlapping problems completely, firstly we extract candidate subjects with a standard span mechanism. Then we present a labeled span mechanism to extract the objects and relations simultaneously, we use the labeled span mechanism to generate labeled spans whose start and end positions indicate the objects, and whose labels correspond to relations of subject and objects. Besides, we design an entity attention mechanism to enhance the information fusion between subject and sentence during extracting objects and relations. We test our method on two public datasets, our method achieves the best performances on these two datasets.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 07:19:39 GMT" } ]
2023-04-07T00:00:00
[ [ "Zhang", "Hao", "" ] ]
TITLE: SpanRE: Entities and Overlapping Relations Extraction Based on Spans and Entity Attention ABSTRACT: Extracting entities and relations is an essential task of information extraction. Triplets extracted from a sentence might overlap with each other. Previous methods either did not address the overlapping issues or solved overlapping issues partially. To tackle triplet overlapping problems completely, firstly we extract candidate subjects with a standard span mechanism. Then we present a labeled span mechanism to extract the objects and relations simultaneously, we use the labeled span mechanism to generate labeled spans whose start and end positions indicate the objects, and whose labels correspond to relations of subject and objects. Besides, we design an entity attention mechanism to enhance the information fusion between subject and sentence during extracting objects and relations. We test our method on two public datasets, our method achieves the best performances on these two datasets.
no_new_dataset
0.711049
2304.02916
Thodoris Kouzelis
Thodoris Kouzelis, Grigoris Bastas, Athanasios Katsamanis and Alexandros Potamianos
Efficient Audio Captioning Transformer with Patchout and Text Guidance
5 pages, 1 figure
null
null
null
cs.SD cs.LG eess.AS
http://creativecommons.org/licenses/by/4.0/
Automated audio captioning is multi-modal translation task that aim to generate textual descriptions for a given audio clip. In this paper we propose a full Transformer architecture that utilizes Patchout as proposed in [1], significantly reducing the computational complexity and avoiding overfitting. The caption generation is partly conditioned on textual AudioSet tags extracted by a pre-trained classification model which is fine-tuned to maximize the semantic similarity between AudioSet labels and ground truth captions. To mitigate the data scarcity problem of Automated Audio Captioning we introduce transfer learning from an upstream audio-related task and an enlarged in-domain dataset. Moreover, we propose a method to apply Mixup augmentation for AAC. Ablation studies are carried out to investigate how Patchout and text guidance contribute to the final performance. The results show that the proposed techniques improve the performance of our system and while reducing the computational complexity. Our proposed method received the Judges Award at the Task6A of DCASE Challenge 2022.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 07:58:27 GMT" } ]
2023-04-07T00:00:00
[ [ "Kouzelis", "Thodoris", "" ], [ "Bastas", "Grigoris", "" ], [ "Katsamanis", "Athanasios", "" ], [ "Potamianos", "Alexandros", "" ] ]
TITLE: Efficient Audio Captioning Transformer with Patchout and Text Guidance ABSTRACT: Automated audio captioning is multi-modal translation task that aim to generate textual descriptions for a given audio clip. In this paper we propose a full Transformer architecture that utilizes Patchout as proposed in [1], significantly reducing the computational complexity and avoiding overfitting. The caption generation is partly conditioned on textual AudioSet tags extracted by a pre-trained classification model which is fine-tuned to maximize the semantic similarity between AudioSet labels and ground truth captions. To mitigate the data scarcity problem of Automated Audio Captioning we introduce transfer learning from an upstream audio-related task and an enlarged in-domain dataset. Moreover, we propose a method to apply Mixup augmentation for AAC. Ablation studies are carried out to investigate how Patchout and text guidance contribute to the final performance. The results show that the proposed techniques improve the performance of our system and while reducing the computational complexity. Our proposed method received the Judges Award at the Task6A of DCASE Challenge 2022.
no_new_dataset
0.70844
2304.02933
Anna Bosman
Hermann Tapamo, Anna Bosman, James Maina and Emile Horak
Convolutional neural networks for crack detection on flexible road pavements
A version of this paper is published in the Proceedings of the 14th International Conference on Soft Computing and Pattern Recognition (SoCPaR 2022). Lecture Notes in Networks and Systems, vol 648. Springer, Cham
null
10.1007/978-3-031-27524-1_19
null
cs.CV cs.AI cs.LG
http://creativecommons.org/licenses/by-nc-nd/4.0/
Flexible road pavements deteriorate primarily due to traffic and adverse environmental conditions. Cracking is the most common deterioration mechanism; the surveying thereof is typically conducted manually using internationally defined classification standards. In South Africa, the use of high-definition video images has been introduced, which allows for safer road surveying. However, surveying is still a tedious manual process. Automation of the detection of defects such as cracks would allow for faster analysis of road networks and potentially reduce human bias and error. This study performs a comparison of six state-of-the-art convolutional neural network models for the purpose of crack detection. The models are pretrained on the ImageNet dataset, and fine-tuned using a new real-world binary crack dataset consisting of 14000 samples. The effects of dataset augmentation are also investigated. Of the six models trained, five achieved accuracy above 97%. The highest recorded accuracy was 98%, achieved by the ResNet and VGG16 models. The dataset is available at the following URL: https://zenodo.org/record/7795975
[ { "version": "v1", "created": "Thu, 6 Apr 2023 08:46:30 GMT" } ]
2023-04-07T00:00:00
[ [ "Tapamo", "Hermann", "" ], [ "Bosman", "Anna", "" ], [ "Maina", "James", "" ], [ "Horak", "Emile", "" ] ]
TITLE: Convolutional neural networks for crack detection on flexible road pavements ABSTRACT: Flexible road pavements deteriorate primarily due to traffic and adverse environmental conditions. Cracking is the most common deterioration mechanism; the surveying thereof is typically conducted manually using internationally defined classification standards. In South Africa, the use of high-definition video images has been introduced, which allows for safer road surveying. However, surveying is still a tedious manual process. Automation of the detection of defects such as cracks would allow for faster analysis of road networks and potentially reduce human bias and error. This study performs a comparison of six state-of-the-art convolutional neural network models for the purpose of crack detection. The models are pretrained on the ImageNet dataset, and fine-tuned using a new real-world binary crack dataset consisting of 14000 samples. The effects of dataset augmentation are also investigated. Of the six models trained, five achieved accuracy above 97%. The highest recorded accuracy was 98%, achieved by the ResNet and VGG16 models. The dataset is available at the following URL: https://zenodo.org/record/7795975
new_dataset
0.716646
2304.02934
Mengmeng Xu
Mengmeng Xu, Mattia Soldan, Jialin Gao, Shuming Liu, Juan-Manuel P\'erez-R\'ua, Bernard Ghanem
Boundary-Denoising for Video Activity Localization
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Video activity localization aims at understanding the semantic content in long untrimmed videos and retrieving actions of interest. The retrieved action with its start and end locations can be used for highlight generation, temporal action detection, etc. Unfortunately, learning the exact boundary location of activities is highly challenging because temporal activities are continuous in time, and there are often no clear-cut transitions between actions. Moreover, the definition of the start and end of events is subjective, which may confuse the model. To alleviate the boundary ambiguity, we propose to study the video activity localization problem from a denoising perspective. Specifically, we propose an encoder-decoder model named DenoiseLoc. During training, a set of action spans is randomly generated from the ground truth with a controlled noise scale. Then we attempt to reverse this process by boundary denoising, allowing the localizer to predict activities with precise boundaries and resulting in faster convergence speed. Experiments show that DenoiseLoc advances %in several video activity understanding tasks. For example, we observe a gain of +12.36% average mAP on QV-Highlights dataset and +1.64% mAP@0.5 on THUMOS'14 dataset over the baseline. Moreover, DenoiseLoc achieves state-of-the-art performance on TACoS and MAD datasets, but with much fewer predictions compared to other current methods.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 08:48:01 GMT" } ]
2023-04-07T00:00:00
[ [ "Xu", "Mengmeng", "" ], [ "Soldan", "Mattia", "" ], [ "Gao", "Jialin", "" ], [ "Liu", "Shuming", "" ], [ "Pérez-Rúa", "Juan-Manuel", "" ], [ "Ghanem", "Bernard", "" ] ]
TITLE: Boundary-Denoising for Video Activity Localization ABSTRACT: Video activity localization aims at understanding the semantic content in long untrimmed videos and retrieving actions of interest. The retrieved action with its start and end locations can be used for highlight generation, temporal action detection, etc. Unfortunately, learning the exact boundary location of activities is highly challenging because temporal activities are continuous in time, and there are often no clear-cut transitions between actions. Moreover, the definition of the start and end of events is subjective, which may confuse the model. To alleviate the boundary ambiguity, we propose to study the video activity localization problem from a denoising perspective. Specifically, we propose an encoder-decoder model named DenoiseLoc. During training, a set of action spans is randomly generated from the ground truth with a controlled noise scale. Then we attempt to reverse this process by boundary denoising, allowing the localizer to predict activities with precise boundaries and resulting in faster convergence speed. Experiments show that DenoiseLoc advances %in several video activity understanding tasks. For example, we observe a gain of +12.36% average mAP on QV-Highlights dataset and +1.64% mAP@0.5 on THUMOS'14 dataset over the baseline. Moreover, DenoiseLoc achieves state-of-the-art performance on TACoS and MAD datasets, but with much fewer predictions compared to other current methods.
no_new_dataset
0.710377
2304.02950
Mingjun Xu
Mingjun Xu, Lingyun Qin, Weijie Chen, Shiliang Pu, Lei Zhang
Multi-view Adversarial Discriminator: Mine the Non-causal Factors for Object Detection in Unseen Domains
CVPR 2023 (Highlight, top 2.5%). Pytorch vs. MindSpore Code at "https://github.com/K2OKOH/MAD"
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Domain shift degrades the performance of object detection models in practical applications. To alleviate the influence of domain shift, plenty of previous work try to decouple and learn the domain-invariant (common) features from source domains via domain adversarial learning (DAL). However, inspired by causal mechanisms, we find that previous methods ignore the implicit insignificant non-causal factors hidden in the common features. This is mainly due to the single-view nature of DAL. In this work, we present an idea to remove non-causal factors from common features by multi-view adversarial training on source domains, because we observe that such insignificant non-causal factors may still be significant in other latent spaces (views) due to the multi-mode structure of data. To summarize, we propose a Multi-view Adversarial Discriminator (MAD) based domain generalization model, consisting of a Spurious Correlations Generator (SCG) that increases the diversity of source domain by random augmentation and a Multi-View Domain Classifier (MVDC) that maps features to multiple latent spaces, such that the non-causal factors are removed and the domain-invariant features are purified. Extensive experiments on six benchmarks show our MAD obtains state-of-the-art performance.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 09:20:28 GMT" } ]
2023-04-07T00:00:00
[ [ "Xu", "Mingjun", "" ], [ "Qin", "Lingyun", "" ], [ "Chen", "Weijie", "" ], [ "Pu", "Shiliang", "" ], [ "Zhang", "Lei", "" ] ]
TITLE: Multi-view Adversarial Discriminator: Mine the Non-causal Factors for Object Detection in Unseen Domains ABSTRACT: Domain shift degrades the performance of object detection models in practical applications. To alleviate the influence of domain shift, plenty of previous work try to decouple and learn the domain-invariant (common) features from source domains via domain adversarial learning (DAL). However, inspired by causal mechanisms, we find that previous methods ignore the implicit insignificant non-causal factors hidden in the common features. This is mainly due to the single-view nature of DAL. In this work, we present an idea to remove non-causal factors from common features by multi-view adversarial training on source domains, because we observe that such insignificant non-causal factors may still be significant in other latent spaces (views) due to the multi-mode structure of data. To summarize, we propose a Multi-view Adversarial Discriminator (MAD) based domain generalization model, consisting of a Spurious Correlations Generator (SCG) that increases the diversity of source domain by random augmentation and a Multi-View Domain Classifier (MVDC) that maps features to multiple latent spaces, such that the non-causal factors are removed and the domain-invariant features are purified. Extensive experiments on six benchmarks show our MAD obtains state-of-the-art performance.
no_new_dataset
0.708616
2304.02982
Ehsan Nowroozi
Ehsan Nowroozi, Yoosef Habibi, Mauro Conti
Spritz-PS: Validation of Synthetic Face Images Using a Large Dataset of Printed Documents
null
null
null
null
cs.CV cs.AI cs.CR cs.CY cs.LG
http://creativecommons.org/licenses/by/4.0/
The capability of doing effective forensic analysis on printed and scanned (PS) images is essential in many applications. PS documents may be used to conceal the artifacts of images which is due to the synthetic nature of images since these artifacts are typically present in manipulated images and the main artifacts in the synthetic images can be removed after the PS. Due to the appeal of Generative Adversarial Networks (GANs), synthetic face images generated with GANs models are difficult to differentiate from genuine human faces and may be used to create counterfeit identities. Additionally, since GANs models do not account for physiological constraints for generating human faces and their impact on human IRISes, distinguishing genuine from synthetic IRISes in the PS scenario becomes extremely difficult. As a result of the lack of large-scale reference IRIS datasets in the PS scenario, we aim at developing a novel dataset to become a standard for Multimedia Forensics (MFs) investigation which is available at [45]. In this paper, we provide a novel dataset made up of a large number of synthetic and natural printed IRISes taken from VIPPrint Printed and Scanned face images. We extracted irises from face images and it is possible that the model due to eyelid occlusion captured the incomplete irises. To fill the missing pixels of extracted iris, we applied techniques to discover the complex link between the iris images. To highlight the problems involved with the evaluation of the dataset's IRIS images, we conducted a large number of analyses employing Siamese Neural Networks to assess the similarities between genuine and synthetic human IRISes, such as ResNet50, Xception, VGG16, and MobileNet-v2. For instance, using the Xception network, we achieved 56.76\% similarity of IRISes for synthetic images and 92.77% similarity of IRISes for real images.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 10:28:34 GMT" } ]
2023-04-07T00:00:00
[ [ "Nowroozi", "Ehsan", "" ], [ "Habibi", "Yoosef", "" ], [ "Conti", "Mauro", "" ] ]
TITLE: Spritz-PS: Validation of Synthetic Face Images Using a Large Dataset of Printed Documents ABSTRACT: The capability of doing effective forensic analysis on printed and scanned (PS) images is essential in many applications. PS documents may be used to conceal the artifacts of images which is due to the synthetic nature of images since these artifacts are typically present in manipulated images and the main artifacts in the synthetic images can be removed after the PS. Due to the appeal of Generative Adversarial Networks (GANs), synthetic face images generated with GANs models are difficult to differentiate from genuine human faces and may be used to create counterfeit identities. Additionally, since GANs models do not account for physiological constraints for generating human faces and their impact on human IRISes, distinguishing genuine from synthetic IRISes in the PS scenario becomes extremely difficult. As a result of the lack of large-scale reference IRIS datasets in the PS scenario, we aim at developing a novel dataset to become a standard for Multimedia Forensics (MFs) investigation which is available at [45]. In this paper, we provide a novel dataset made up of a large number of synthetic and natural printed IRISes taken from VIPPrint Printed and Scanned face images. We extracted irises from face images and it is possible that the model due to eyelid occlusion captured the incomplete irises. To fill the missing pixels of extracted iris, we applied techniques to discover the complex link between the iris images. To highlight the problems involved with the evaluation of the dataset's IRIS images, we conducted a large number of analyses employing Siamese Neural Networks to assess the similarities between genuine and synthetic human IRISes, such as ResNet50, Xception, VGG16, and MobileNet-v2. For instance, using the Xception network, we achieved 56.76\% similarity of IRISes for synthetic images and 92.77% similarity of IRISes for real images.
new_dataset
0.717791
2304.02986
Thibault Gauthier
Thibault Gauthier, Chad E. Brown, Mikolas Janota, Josef Urban
A Mathematical Benchmark for Inductive Theorem Provers
null
null
null
null
cs.LO
http://creativecommons.org/licenses/by-nc-sa/4.0/
We present a benchmark of 29687 problems derived from the On-Line Encyclopedia of Integer Sequences (OEIS). Each problem expresses the equivalence of two syntactically different programs generating the same OEIS sequence. Such programs were conjectured by a learning-guided synthesis system using a language with looping operators. The operators implement recursion, and thus many of the proofs require induction on natural numbers. The benchmark contains problems of varying difficulty from a wide area of mathematical domains. We believe that these characteristics will make it an effective judge for the progress of inductive theorem provers in this domain for years to come.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 10:41:51 GMT" } ]
2023-04-07T00:00:00
[ [ "Gauthier", "Thibault", "" ], [ "Brown", "Chad E.", "" ], [ "Janota", "Mikolas", "" ], [ "Urban", "Josef", "" ] ]
TITLE: A Mathematical Benchmark for Inductive Theorem Provers ABSTRACT: We present a benchmark of 29687 problems derived from the On-Line Encyclopedia of Integer Sequences (OEIS). Each problem expresses the equivalence of two syntactically different programs generating the same OEIS sequence. Such programs were conjectured by a learning-guided synthesis system using a language with looping operators. The operators implement recursion, and thus many of the proofs require induction on natural numbers. The benchmark contains problems of varying difficulty from a wide area of mathematical domains. We believe that these characteristics will make it an effective judge for the progress of inductive theorem provers in this domain for years to come.
new_dataset
0.63114
2304.02991
Adriano Cardace
Adriano Cardace, Pierluigi Zama Ramirez, Samuele Salti, Luigi Di Stefano
Exploiting the Complementarity of 2D and 3D Networks to Address Domain-Shift in 3D Semantic Segmentation
Accepted at the CVPR2023 Workshop on Autonomous Driving (WAD)
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
3D semantic segmentation is a critical task in many real-world applications, such as autonomous driving, robotics, and mixed reality. However, the task is extremely challenging due to ambiguities coming from the unstructured, sparse, and uncolored nature of the 3D point clouds. A possible solution is to combine the 3D information with others coming from sensors featuring a different modality, such as RGB cameras. Recent multi-modal 3D semantic segmentation networks exploit these modalities relying on two branches that process the 2D and 3D information independently, striving to maintain the strength of each modality. In this work, we first explain why this design choice is effective and then show how it can be improved to make the multi-modal semantic segmentation more robust to domain shift. Our surprisingly simple contribution achieves state-of-the-art performances on four popular multi-modal unsupervised domain adaptation benchmarks, as well as better results in a domain generalization scenario.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 10:59:43 GMT" } ]
2023-04-07T00:00:00
[ [ "Cardace", "Adriano", "" ], [ "Ramirez", "Pierluigi Zama", "" ], [ "Salti", "Samuele", "" ], [ "Di Stefano", "Luigi", "" ] ]
TITLE: Exploiting the Complementarity of 2D and 3D Networks to Address Domain-Shift in 3D Semantic Segmentation ABSTRACT: 3D semantic segmentation is a critical task in many real-world applications, such as autonomous driving, robotics, and mixed reality. However, the task is extremely challenging due to ambiguities coming from the unstructured, sparse, and uncolored nature of the 3D point clouds. A possible solution is to combine the 3D information with others coming from sensors featuring a different modality, such as RGB cameras. Recent multi-modal 3D semantic segmentation networks exploit these modalities relying on two branches that process the 2D and 3D information independently, striving to maintain the strength of each modality. In this work, we first explain why this design choice is effective and then show how it can be improved to make the multi-modal semantic segmentation more robust to domain shift. Our surprisingly simple contribution achieves state-of-the-art performances on four popular multi-modal unsupervised domain adaptation benchmarks, as well as better results in a domain generalization scenario.
no_new_dataset
0.712045
2304.03006
James Calo
James Calo and Benny Lo
IoT Federated Blockchain Learning at the Edge
Submitted to IEEE EMBC 2023
null
null
null
cs.LG cs.CR
http://creativecommons.org/licenses/by/4.0/
IoT devices are sorely underutilized in the medical field, especially within machine learning for medicine, yet they offer unrivaled benefits. IoT devices are low-cost, energy-efficient, small and intelligent devices. In this paper, we propose a distributed federated learning framework for IoT devices, more specifically for IoMT (Internet of Medical Things), using blockchain to allow for a decentralized scheme improving privacy and efficiency over a centralized system; this allows us to move from the cloud-based architectures, that are prevalent, to the edge. The system is designed for three paradigms: 1) Training neural networks on IoT devices to allow for collaborative training of a shared model whilst decoupling the learning from the dataset to ensure privacy. Training is performed in an online manner simultaneously amongst all participants, allowing for the training of actual data that may not have been present in a dataset collected in the traditional way and dynamically adapt the system whilst it is being trained. 2) Training of an IoMT system in a fully private manner such as to mitigate the issue with confidentiality of medical data and to build robust, and potentially bespoke, models where not much, if any, data exists. 3) Distribution of the actual network training, something federated learning itself does not do, to allow hospitals, for example, to utilize their spare computing resources to train network models.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 11:32:40 GMT" } ]
2023-04-07T00:00:00
[ [ "Calo", "James", "" ], [ "Lo", "Benny", "" ] ]
TITLE: IoT Federated Blockchain Learning at the Edge ABSTRACT: IoT devices are sorely underutilized in the medical field, especially within machine learning for medicine, yet they offer unrivaled benefits. IoT devices are low-cost, energy-efficient, small and intelligent devices. In this paper, we propose a distributed federated learning framework for IoT devices, more specifically for IoMT (Internet of Medical Things), using blockchain to allow for a decentralized scheme improving privacy and efficiency over a centralized system; this allows us to move from the cloud-based architectures, that are prevalent, to the edge. The system is designed for three paradigms: 1) Training neural networks on IoT devices to allow for collaborative training of a shared model whilst decoupling the learning from the dataset to ensure privacy. Training is performed in an online manner simultaneously amongst all participants, allowing for the training of actual data that may not have been present in a dataset collected in the traditional way and dynamically adapt the system whilst it is being trained. 2) Training of an IoMT system in a fully private manner such as to mitigate the issue with confidentiality of medical data and to build robust, and potentially bespoke, models where not much, if any, data exists. 3) Distribution of the actual network training, something federated learning itself does not do, to allow hospitals, for example, to utilize their spare computing resources to train network models.
no_new_dataset
0.711224
2304.03022
Chen Li
Chen Li, Yixiao Ge, Jiayong Mao, Dian Li, Ying Shan
TagGPT: Large Language Models are Zero-shot Multimodal Taggers
13 pages, 6 figures
null
null
null
cs.IR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Tags are pivotal in facilitating the effective distribution of multimedia content in various applications in the contemporary Internet era, such as search engines and recommendation systems. Recently, large language models (LLMs) have demonstrated impressive capabilities across a wide range of tasks. In this work, we propose TagGPT, a fully automated system capable of tag extraction and multimodal tagging in a completely zero-shot fashion. Our core insight is that, through elaborate prompt engineering, LLMs are able to extract and reason about proper tags given textual clues of multimodal data, e.g., OCR, ASR, title, etc. Specifically, to automatically build a high-quality tag set that reflects user intent and interests for a specific application, TagGPT predicts large-scale candidate tags from a series of raw data via prompting LLMs, filtered with frequency and semantics. Given a new entity that needs tagging for distribution, TagGPT introduces two alternative options for zero-shot tagging, i.e., a generative method with late semantic matching with the tag set, and another selective method with early matching in prompts. It is well noticed that TagGPT provides a system-level solution based on a modular framework equipped with a pre-trained LLM (GPT-3.5 used here) and a sentence embedding model (SimCSE used here), which can be seamlessly replaced with any more advanced one you want. TagGPT is applicable for various modalities of data in modern social media and showcases strong generalization ability to a wide range of applications. We evaluate TagGPT on publicly available datasets, i.e., Kuaishou and Food.com, and demonstrate the effectiveness of TagGPT compared to existing hashtags and off-the-shelf taggers. Project page: https://github.com/TencentARC/TagGPT.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 12:17:46 GMT" } ]
2023-04-07T00:00:00
[ [ "Li", "Chen", "" ], [ "Ge", "Yixiao", "" ], [ "Mao", "Jiayong", "" ], [ "Li", "Dian", "" ], [ "Shan", "Ying", "" ] ]
TITLE: TagGPT: Large Language Models are Zero-shot Multimodal Taggers ABSTRACT: Tags are pivotal in facilitating the effective distribution of multimedia content in various applications in the contemporary Internet era, such as search engines and recommendation systems. Recently, large language models (LLMs) have demonstrated impressive capabilities across a wide range of tasks. In this work, we propose TagGPT, a fully automated system capable of tag extraction and multimodal tagging in a completely zero-shot fashion. Our core insight is that, through elaborate prompt engineering, LLMs are able to extract and reason about proper tags given textual clues of multimodal data, e.g., OCR, ASR, title, etc. Specifically, to automatically build a high-quality tag set that reflects user intent and interests for a specific application, TagGPT predicts large-scale candidate tags from a series of raw data via prompting LLMs, filtered with frequency and semantics. Given a new entity that needs tagging for distribution, TagGPT introduces two alternative options for zero-shot tagging, i.e., a generative method with late semantic matching with the tag set, and another selective method with early matching in prompts. It is well noticed that TagGPT provides a system-level solution based on a modular framework equipped with a pre-trained LLM (GPT-3.5 used here) and a sentence embedding model (SimCSE used here), which can be seamlessly replaced with any more advanced one you want. TagGPT is applicable for various modalities of data in modern social media and showcases strong generalization ability to a wide range of applications. We evaluate TagGPT on publicly available datasets, i.e., Kuaishou and Food.com, and demonstrate the effectiveness of TagGPT compared to existing hashtags and off-the-shelf taggers. Project page: https://github.com/TencentARC/TagGPT.
no_new_dataset
0.70724
2304.03064
Khloud Al Jallad
Rahmeh Abou Zafra, Lana Ahmad Abdullah, Rouaa Alaraj, Rasha Albezreh, Tarek Barhoum, Khloud Al Jallad
An experimental study in Real-time Facial Emotion Recognition on new 3RL dataset
null
Journal of Current Trends in Computer Science Research, 2(2), 68-76 (2023)
10.33140/JCTCSR
null
cs.CV cs.AI cs.HC cs.LG
http://creativecommons.org/licenses/by/4.0/
Although real-time facial emotion recognition is a hot topic research domain in the field of human-computer interaction, state-of the-art available datasets still suffer from various problems, such as some unrelated photos such as document photos, unbalanced numbers of photos in each class, and misleading images that can negatively affect correct classification. The 3RL dataset was created, which contains approximately 24K images and will be publicly available, to overcome previously available dataset problems. The 3RL dataset is labelled with five basic emotions: happiness, fear, sadness, disgust, and anger. Moreover, we compared the 3RL dataset with other famous state-of-the-art datasets (FER dataset, CK+ dataset), and we applied the most commonly used algorithms in previous works, SVM and CNN. The results show a noticeable improvement in generalization on the 3RL dataset. Experiments have shown an accuracy of up to 91.4% on 3RL dataset using CNN where results on FER2013, CK+ are, respectively (approximately from 60% to 85%).
[ { "version": "v1", "created": "Thu, 6 Apr 2023 13:29:11 GMT" } ]
2023-04-07T00:00:00
[ [ "Zafra", "Rahmeh Abou", "" ], [ "Abdullah", "Lana Ahmad", "" ], [ "Alaraj", "Rouaa", "" ], [ "Albezreh", "Rasha", "" ], [ "Barhoum", "Tarek", "" ], [ "Jallad", "Khloud Al", "" ] ]
TITLE: An experimental study in Real-time Facial Emotion Recognition on new 3RL dataset ABSTRACT: Although real-time facial emotion recognition is a hot topic research domain in the field of human-computer interaction, state-of the-art available datasets still suffer from various problems, such as some unrelated photos such as document photos, unbalanced numbers of photos in each class, and misleading images that can negatively affect correct classification. The 3RL dataset was created, which contains approximately 24K images and will be publicly available, to overcome previously available dataset problems. The 3RL dataset is labelled with five basic emotions: happiness, fear, sadness, disgust, and anger. Moreover, we compared the 3RL dataset with other famous state-of-the-art datasets (FER dataset, CK+ dataset), and we applied the most commonly used algorithms in previous works, SVM and CNN. The results show a noticeable improvement in generalization on the 3RL dataset. Experiments have shown an accuracy of up to 91.4% on 3RL dataset using CNN where results on FER2013, CK+ are, respectively (approximately from 60% to 85%).
new_dataset
0.718943
2304.03096
Edric Tam
Edric Tam, David Dunson
Spectral Gap Regularization of Neural Networks
This is a journal extension of the ICML conference paper by Tam and Dunson (2020), arXiv:2003.00992
null
null
null
stat.ML cs.LG
http://creativecommons.org/licenses/by/4.0/
We introduce Fiedler regularization, a novel approach for regularizing neural networks that utilizes spectral/graphical information. Existing regularization methods often focus on penalizing weights in a global/uniform manner that ignores the connectivity structure of the neural network. We propose to use the Fiedler value of the neural network's underlying graph as a tool for regularization. We provide theoretical motivation for this approach via spectral graph theory. We demonstrate several useful properties of the Fiedler value that make it useful as a regularization tool. We provide an approximate, variational approach for faster computation during training. We provide an alternative formulation of this framework in the form of a structurally weighted $\text{L}_1$ penalty, thus linking our approach to sparsity induction. We provide uniform generalization error bounds for Fiedler regularization via a Rademacher complexity analysis. We performed experiments on datasets that compare Fiedler regularization with classical regularization methods such as dropout and weight decay. Results demonstrate the efficacy of Fiedler regularization. This is a journal extension of the conference paper by Tam and Dunson (2020).
[ { "version": "v1", "created": "Thu, 6 Apr 2023 14:23:40 GMT" } ]
2023-04-07T00:00:00
[ [ "Tam", "Edric", "" ], [ "Dunson", "David", "" ] ]
TITLE: Spectral Gap Regularization of Neural Networks ABSTRACT: We introduce Fiedler regularization, a novel approach for regularizing neural networks that utilizes spectral/graphical information. Existing regularization methods often focus on penalizing weights in a global/uniform manner that ignores the connectivity structure of the neural network. We propose to use the Fiedler value of the neural network's underlying graph as a tool for regularization. We provide theoretical motivation for this approach via spectral graph theory. We demonstrate several useful properties of the Fiedler value that make it useful as a regularization tool. We provide an approximate, variational approach for faster computation during training. We provide an alternative formulation of this framework in the form of a structurally weighted $\text{L}_1$ penalty, thus linking our approach to sparsity induction. We provide uniform generalization error bounds for Fiedler regularization via a Rademacher complexity analysis. We performed experiments on datasets that compare Fiedler regularization with classical regularization methods such as dropout and weight decay. Results demonstrate the efficacy of Fiedler regularization. This is a journal extension of the conference paper by Tam and Dunson (2020).
no_new_dataset
0.708023
2304.03098
Matteo Muffo
Matteo Muffo, Roberto Tedesco, Licia Sbattella and Vincenzo Scotti
Static Fuzzy Bag-of-Words: a lightweight sentence embedding algorithm
9 pages, 2 figures
Proceedings of the 4th International Conference on Natural Language and Speech Processing (ICNLSP 2021)
null
null
cs.CL stat.ML
http://creativecommons.org/licenses/by/4.0/
The introduction of embedding techniques has pushed forward significantly the Natural Language Processing field. Many of the proposed solutions have been presented for word-level encoding; anyhow, in the last years, new mechanism to treat information at an higher level of aggregation, like at sentence- and document-level, have emerged. With this work we address specifically the sentence embeddings problem, presenting the Static Fuzzy Bag-of-Word model. Our model is a refinement of the Fuzzy Bag-of-Words approach, providing sentence embeddings with a predefined dimension. SFBoW provides competitive performances in Semantic Textual Similarity benchmarks, while requiring low computational resources.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 14:25:46 GMT" } ]
2023-04-07T00:00:00
[ [ "Muffo", "Matteo", "" ], [ "Tedesco", "Roberto", "" ], [ "Sbattella", "Licia", "" ], [ "Scotti", "Vincenzo", "" ] ]
TITLE: Static Fuzzy Bag-of-Words: a lightweight sentence embedding algorithm ABSTRACT: The introduction of embedding techniques has pushed forward significantly the Natural Language Processing field. Many of the proposed solutions have been presented for word-level encoding; anyhow, in the last years, new mechanism to treat information at an higher level of aggregation, like at sentence- and document-level, have emerged. With this work we address specifically the sentence embeddings problem, presenting the Static Fuzzy Bag-of-Word model. Our model is a refinement of the Fuzzy Bag-of-Words approach, providing sentence embeddings with a predefined dimension. SFBoW provides competitive performances in Semantic Textual Similarity benchmarks, while requiring low computational resources.
no_new_dataset
0.711663
2304.03101
Lydia Anette Sch\"onpflug
Lydia A. Schoenpflug, Maxime W. Lafarge, Anja L. Frei, Viktor H. Koelzer
Multi-task learning for tissue segmentation and tumor detection in colorectal cancer histology slides
null
null
null
null
eess.IV cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Automating tissue segmentation and tumor detection in histopathology images of colorectal cancer (CRC) is an enabler for faster diagnostic pathology workflows. At the same time it is a challenging task due to low availability of public annotated datasets and high variability of image appearance. The semi-supervised learning for CRC detection (SemiCOL) challenge 2023 provides partially annotated data to encourage the development of automated solutions for tissue segmentation and tumor detection. We propose a U-Net based multi-task model combined with channel-wise and image-statistics-based color augmentations, as well as test-time augmentation, as a candidate solution to the SemiCOL challenge. Our approach achieved a multi-task Dice score of .8655 (Arm 1) and .8515 (Arm 2) for tissue segmentation and AUROC of .9725 (Arm 1) and 0.9750 (Arm 2) for tumor detection on the challenge validation set. The source code for our approach is made publicly available at https://github.com/lely475/CTPLab_SemiCOL2023.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 14:26:41 GMT" } ]
2023-04-07T00:00:00
[ [ "Schoenpflug", "Lydia A.", "" ], [ "Lafarge", "Maxime W.", "" ], [ "Frei", "Anja L.", "" ], [ "Koelzer", "Viktor H.", "" ] ]
TITLE: Multi-task learning for tissue segmentation and tumor detection in colorectal cancer histology slides ABSTRACT: Automating tissue segmentation and tumor detection in histopathology images of colorectal cancer (CRC) is an enabler for faster diagnostic pathology workflows. At the same time it is a challenging task due to low availability of public annotated datasets and high variability of image appearance. The semi-supervised learning for CRC detection (SemiCOL) challenge 2023 provides partially annotated data to encourage the development of automated solutions for tissue segmentation and tumor detection. We propose a U-Net based multi-task model combined with channel-wise and image-statistics-based color augmentations, as well as test-time augmentation, as a candidate solution to the SemiCOL challenge. Our approach achieved a multi-task Dice score of .8655 (Arm 1) and .8515 (Arm 2) for tissue segmentation and AUROC of .9725 (Arm 1) and 0.9750 (Arm 2) for tumor detection on the challenge validation set. The source code for our approach is made publicly available at https://github.com/lely475/CTPLab_SemiCOL2023.
no_new_dataset
0.708648
2304.03112
Andreea Iana
Andreea Iana, Goran Glava\v{s}, and Heiko Paulheim
Simplifying Content-Based Neural News Recommendation: On User Modeling and Training Objectives
Accepted at the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2023)
null
null
null
cs.IR
http://creativecommons.org/licenses/by/4.0/
The advent of personalized news recommendation has given rise to increasingly complex recommender architectures. Most neural news recommenders rely on user click behavior and typically introduce dedicated user encoders that aggregate the content of clicked news into user embeddings (early fusion). These models are predominantly trained with standard point-wise classification objectives. The existing body of work exhibits two main shortcomings: (1) despite general design homogeneity, direct comparisons between models are hindered by varying evaluation datasets and protocols; (2) it leaves alternative model designs and training objectives vastly unexplored. In this work, we present a unified framework for news recommendation, allowing for a systematic and fair comparison of news recommenders across several crucial design dimensions: (i) candidate-awareness in user modeling, (ii) click behavior fusion, and (iii) training objectives. Our findings challenge the status quo in neural news recommendation. We show that replacing sizable user encoders with parameter-efficient dot products between candidate and clicked news embeddings (late fusion) often yields substantial performance gains. Moreover, our results render contrastive training a viable alternative to point-wise classification objectives.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 14:40:01 GMT" } ]
2023-04-07T00:00:00
[ [ "Iana", "Andreea", "" ], [ "Glavaš", "Goran", "" ], [ "Paulheim", "Heiko", "" ] ]
TITLE: Simplifying Content-Based Neural News Recommendation: On User Modeling and Training Objectives ABSTRACT: The advent of personalized news recommendation has given rise to increasingly complex recommender architectures. Most neural news recommenders rely on user click behavior and typically introduce dedicated user encoders that aggregate the content of clicked news into user embeddings (early fusion). These models are predominantly trained with standard point-wise classification objectives. The existing body of work exhibits two main shortcomings: (1) despite general design homogeneity, direct comparisons between models are hindered by varying evaluation datasets and protocols; (2) it leaves alternative model designs and training objectives vastly unexplored. In this work, we present a unified framework for news recommendation, allowing for a systematic and fair comparison of news recommenders across several crucial design dimensions: (i) candidate-awareness in user modeling, (ii) click behavior fusion, and (iii) training objectives. Our findings challenge the status quo in neural news recommendation. We show that replacing sizable user encoders with parameter-efficient dot products between candidate and clicked news embeddings (late fusion) often yields substantial performance gains. Moreover, our results render contrastive training a viable alternative to point-wise classification objectives.
no_new_dataset
0.710465
2304.03135
Mengyin Liu
Mengyin Liu, Jie Jiang, Chao Zhu, Xu-Cheng Yin
VLPD: Context-Aware Pedestrian Detection via Vision-Language Semantic Self-Supervision
Accepted by CVPR 2023
null
null
null
cs.CV cs.AI cs.MM
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Detecting pedestrians accurately in urban scenes is significant for realistic applications like autonomous driving or video surveillance. However, confusing human-like objects often lead to wrong detections, and small scale or heavily occluded pedestrians are easily missed due to their unusual appearances. To address these challenges, only object regions are inadequate, thus how to fully utilize more explicit and semantic contexts becomes a key problem. Meanwhile, previous context-aware pedestrian detectors either only learn latent contexts with visual clues, or need laborious annotations to obtain explicit and semantic contexts. Therefore, we propose in this paper a novel approach via Vision-Language semantic self-supervision for context-aware Pedestrian Detection (VLPD) to model explicitly semantic contexts without any extra annotations. Firstly, we propose a self-supervised Vision-Language Semantic (VLS) segmentation method, which learns both fully-supervised pedestrian detection and contextual segmentation via self-generated explicit labels of semantic classes by vision-language models. Furthermore, a self-supervised Prototypical Semantic Contrastive (PSC) learning method is proposed to better discriminate pedestrians and other classes, based on more explicit and semantic contexts obtained from VLS. Extensive experiments on popular benchmarks show that our proposed VLPD achieves superior performances over the previous state-of-the-arts, particularly under challenging circumstances like small scale and heavy occlusion. Code is available at https://github.com/lmy98129/VLPD.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 15:16:29 GMT" } ]
2023-04-07T00:00:00
[ [ "Liu", "Mengyin", "" ], [ "Jiang", "Jie", "" ], [ "Zhu", "Chao", "" ], [ "Yin", "Xu-Cheng", "" ] ]
TITLE: VLPD: Context-Aware Pedestrian Detection via Vision-Language Semantic Self-Supervision ABSTRACT: Detecting pedestrians accurately in urban scenes is significant for realistic applications like autonomous driving or video surveillance. However, confusing human-like objects often lead to wrong detections, and small scale or heavily occluded pedestrians are easily missed due to their unusual appearances. To address these challenges, only object regions are inadequate, thus how to fully utilize more explicit and semantic contexts becomes a key problem. Meanwhile, previous context-aware pedestrian detectors either only learn latent contexts with visual clues, or need laborious annotations to obtain explicit and semantic contexts. Therefore, we propose in this paper a novel approach via Vision-Language semantic self-supervision for context-aware Pedestrian Detection (VLPD) to model explicitly semantic contexts without any extra annotations. Firstly, we propose a self-supervised Vision-Language Semantic (VLS) segmentation method, which learns both fully-supervised pedestrian detection and contextual segmentation via self-generated explicit labels of semantic classes by vision-language models. Furthermore, a self-supervised Prototypical Semantic Contrastive (PSC) learning method is proposed to better discriminate pedestrians and other classes, based on more explicit and semantic contexts obtained from VLS. Extensive experiments on popular benchmarks show that our proposed VLPD achieves superior performances over the previous state-of-the-arts, particularly under challenging circumstances like small scale and heavy occlusion. Code is available at https://github.com/lmy98129/VLPD.
no_new_dataset
0.712188
2304.03140
Changsheng Lu
Changsheng Lu, Hao Zhu, Piotr Koniusz
From Saliency to DINO: Saliency-guided Vision Transformer for Few-shot Keypoint Detection
15 pages, 10 figures
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Unlike current deep keypoint detectors that are trained to recognize limited number of body parts, few-shot keypoint detection (FSKD) attempts to localize any keypoints, including novel or base keypoints, depending on the reference samples. FSKD requires the semantically meaningful relations for keypoint similarity learning to overcome the ubiquitous noise and ambiguous local patterns. One rescue comes with vision transformer (ViT) as it captures long-range relations well. However, ViT may model irrelevant features outside of the region of interest due to the global attention matrix, thus degrading similarity learning between support and query features. In this paper, we present a novel saliency-guided vision transformer, dubbed SalViT, for few-shot keypoint detection. Our SalViT enjoys a uniquely designed masked self-attention and a morphology learner, where the former introduces saliency map as a soft mask to constrain the self-attention on foregrounds, while the latter leverages the so-called power normalization to adjust morphology of saliency map, realizing ``dynamically changing receptive field''. Moreover, as salinecy detectors add computations, we show that attentive masks of DINO transformer can replace saliency. On top of SalViT, we also investigate i) transductive FSKD that enhances keypoint representations with unlabelled data and ii) FSKD under occlusions. We show that our model performs well on five public datasets and achieves ~10% PCK higher than the normally trained model under severe occlusions.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 15:22:34 GMT" } ]
2023-04-07T00:00:00
[ [ "Lu", "Changsheng", "" ], [ "Zhu", "Hao", "" ], [ "Koniusz", "Piotr", "" ] ]
TITLE: From Saliency to DINO: Saliency-guided Vision Transformer for Few-shot Keypoint Detection ABSTRACT: Unlike current deep keypoint detectors that are trained to recognize limited number of body parts, few-shot keypoint detection (FSKD) attempts to localize any keypoints, including novel or base keypoints, depending on the reference samples. FSKD requires the semantically meaningful relations for keypoint similarity learning to overcome the ubiquitous noise and ambiguous local patterns. One rescue comes with vision transformer (ViT) as it captures long-range relations well. However, ViT may model irrelevant features outside of the region of interest due to the global attention matrix, thus degrading similarity learning between support and query features. In this paper, we present a novel saliency-guided vision transformer, dubbed SalViT, for few-shot keypoint detection. Our SalViT enjoys a uniquely designed masked self-attention and a morphology learner, where the former introduces saliency map as a soft mask to constrain the self-attention on foregrounds, while the latter leverages the so-called power normalization to adjust morphology of saliency map, realizing ``dynamically changing receptive field''. Moreover, as salinecy detectors add computations, we show that attentive masks of DINO transformer can replace saliency. On top of SalViT, we also investigate i) transductive FSKD that enhances keypoint representations with unlabelled data and ii) FSKD under occlusions. We show that our model performs well on five public datasets and achieves ~10% PCK higher than the normally trained model under severe occlusions.
no_new_dataset
0.709598
2304.03147
Jia-Hong Huang
Jia-Hong Huang, Modar Alfadly, Bernard Ghanem, Marcel Worring
Improving Visual Question Answering Models through Robustness Analysis and In-Context Learning with a Chain of Basic Questions
28 pages
null
null
null
cs.CV cs.AI
http://creativecommons.org/licenses/by/4.0/
Deep neural networks have been critical in the task of Visual Question Answering (VQA), with research traditionally focused on improving model accuracy. Recently, however, there has been a trend towards evaluating the robustness of these models against adversarial attacks. This involves assessing the accuracy of VQA models under increasing levels of noise in the input, which can target either the image or the proposed query question, dubbed the main question. However, there is currently a lack of proper analysis of this aspect of VQA. This work proposes a new method that utilizes semantically related questions, referred to as basic questions, acting as noise to evaluate the robustness of VQA models. It is hypothesized that as the similarity of a basic question to the main question decreases, the level of noise increases. To generate a reasonable noise level for a given main question, a pool of basic questions is ranked based on their similarity to the main question, and this ranking problem is cast as a LASSO optimization problem. Additionally, this work proposes a novel robustness measure, R_score, and two basic question datasets to standardize the analysis of VQA model robustness. The experimental results demonstrate that the proposed evaluation method effectively analyzes the robustness of VQA models. Moreover, the experiments show that in-context learning with a chain of basic questions can enhance model accuracy.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 15:32:35 GMT" } ]
2023-04-07T00:00:00
[ [ "Huang", "Jia-Hong", "" ], [ "Alfadly", "Modar", "" ], [ "Ghanem", "Bernard", "" ], [ "Worring", "Marcel", "" ] ]
TITLE: Improving Visual Question Answering Models through Robustness Analysis and In-Context Learning with a Chain of Basic Questions ABSTRACT: Deep neural networks have been critical in the task of Visual Question Answering (VQA), with research traditionally focused on improving model accuracy. Recently, however, there has been a trend towards evaluating the robustness of these models against adversarial attacks. This involves assessing the accuracy of VQA models under increasing levels of noise in the input, which can target either the image or the proposed query question, dubbed the main question. However, there is currently a lack of proper analysis of this aspect of VQA. This work proposes a new method that utilizes semantically related questions, referred to as basic questions, acting as noise to evaluate the robustness of VQA models. It is hypothesized that as the similarity of a basic question to the main question decreases, the level of noise increases. To generate a reasonable noise level for a given main question, a pool of basic questions is ranked based on their similarity to the main question, and this ranking problem is cast as a LASSO optimization problem. Additionally, this work proposes a novel robustness measure, R_score, and two basic question datasets to standardize the analysis of VQA model robustness. The experimental results demonstrate that the proposed evaluation method effectively analyzes the robustness of VQA models. Moreover, the experiments show that in-context learning with a chain of basic questions can enhance model accuracy.
no_new_dataset
0.707367
2304.03148
Jaeyoun You
Jaeyoun You, Jinhan Choi, Ho-Jae Shin, Bongwon Suh
The Eyes Have It!: Using Human-Selected Features for Predicting Athletes' Performance
null
null
null
null
cs.HC
http://creativecommons.org/licenses/by/4.0/
Predicting athletes' performance has relied mostly on statistical data. Besides the traditional data, various types of data, including video, have become available. However, it is challenging to use them for deep learning, especially when the size of the athletes' dataset is small. This research proposes a feature-selection strategy based on the criteria used by insightful people, which could improve ML performance. Our ML model employs features selected by people who correctly evaluated the athletes' future performance. We tested out a strategy to predict the LPGA players' next day performance using their interview video. We asked study participants to predict the players' next day score after watching the interviews and asked why. Using combined features of the facial landmarks' movements, derived from the participants, and meta-data showed a better F1-score than using each feature separately. This study suggests that the human-in-the-loop model could improve algorithms' performance with small-dataset.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 15:32:41 GMT" } ]
2023-04-07T00:00:00
[ [ "You", "Jaeyoun", "" ], [ "Choi", "Jinhan", "" ], [ "Shin", "Ho-Jae", "" ], [ "Suh", "Bongwon", "" ] ]
TITLE: The Eyes Have It!: Using Human-Selected Features for Predicting Athletes' Performance ABSTRACT: Predicting athletes' performance has relied mostly on statistical data. Besides the traditional data, various types of data, including video, have become available. However, it is challenging to use them for deep learning, especially when the size of the athletes' dataset is small. This research proposes a feature-selection strategy based on the criteria used by insightful people, which could improve ML performance. Our ML model employs features selected by people who correctly evaluated the athletes' future performance. We tested out a strategy to predict the LPGA players' next day performance using their interview video. We asked study participants to predict the players' next day score after watching the interviews and asked why. Using combined features of the facial landmarks' movements, derived from the participants, and meta-data showed a better F1-score than using each feature separately. This study suggests that the human-in-the-loop model could improve algorithms' performance with small-dataset.
no_new_dataset
0.710069
2304.03153
Lei Wang
Lei Wang, Ee-Peng Lim
Zero-Shot Next-Item Recommendation using Large Pretrained Language Models
Technical Report
null
null
null
cs.IR cs.CL
http://creativecommons.org/licenses/by/4.0/
Large language models (LLMs) have achieved impressive zero-shot performance in various natural language processing (NLP) tasks, demonstrating their capabilities for inference without training examples. Despite their success, no research has yet explored the potential of LLMs to perform next-item recommendations in the zero-shot setting. We have identified two major challenges that must be addressed to enable LLMs to act effectively as recommenders. First, the recommendation space can be extremely large for LLMs, and LLMs do not know about the target user's past interacted items and preferences. To address this gap, we propose a prompting strategy called Zero-Shot Next-Item Recommendation (NIR) prompting that directs LLMs to make next-item recommendations. Specifically, the NIR-based strategy involves using an external module to generate candidate items based on user-filtering or item-filtering. Our strategy incorporates a 3-step prompting that guides GPT-3 to carry subtasks that capture the user's preferences, select representative previously watched movies, and recommend a ranked list of 10 movies. We evaluate the proposed approach using GPT-3 on MovieLens 100K dataset and show that it achieves strong zero-shot performance, even outperforming some strong sequential recommendation models trained on the entire training dataset. These promising results highlight the ample research opportunities to use LLMs as recommenders. The code can be found at https://github.com/AGI-Edgerunners/LLM-Next-Item-Rec.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 15:35:11 GMT" } ]
2023-04-07T00:00:00
[ [ "Wang", "Lei", "" ], [ "Lim", "Ee-Peng", "" ] ]
TITLE: Zero-Shot Next-Item Recommendation using Large Pretrained Language Models ABSTRACT: Large language models (LLMs) have achieved impressive zero-shot performance in various natural language processing (NLP) tasks, demonstrating their capabilities for inference without training examples. Despite their success, no research has yet explored the potential of LLMs to perform next-item recommendations in the zero-shot setting. We have identified two major challenges that must be addressed to enable LLMs to act effectively as recommenders. First, the recommendation space can be extremely large for LLMs, and LLMs do not know about the target user's past interacted items and preferences. To address this gap, we propose a prompting strategy called Zero-Shot Next-Item Recommendation (NIR) prompting that directs LLMs to make next-item recommendations. Specifically, the NIR-based strategy involves using an external module to generate candidate items based on user-filtering or item-filtering. Our strategy incorporates a 3-step prompting that guides GPT-3 to carry subtasks that capture the user's preferences, select representative previously watched movies, and recommend a ranked list of 10 movies. We evaluate the proposed approach using GPT-3 on MovieLens 100K dataset and show that it achieves strong zero-shot performance, even outperforming some strong sequential recommendation models trained on the entire training dataset. These promising results highlight the ample research opportunities to use LLMs as recommenders. The code can be found at https://github.com/AGI-Edgerunners/LLM-Next-Item-Rec.
no_new_dataset
0.710622
2304.03156
Kshitij Agrawal
Sri Charan Kattamuru, Kshitij Agrawal, Shyam Prasad Adhikari, Abhishek Bose, Hemant Misra
Patch-wise Features for Blur Image Classification
Accepted in AIMLSystems2022
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Images captured through smartphone cameras often suffer from degradation, blur being one of the major ones, posing a challenge in processing these images for downstream tasks. In this paper we propose low-compute lightweight patch-wise features for image quality assessment. Using our method we can discriminate between blur vs sharp image degradation. To this end, we train a decision-tree based XGBoost model on various intuitive image features like gray level variance, first and second order gradients, texture features like local binary patterns. Experiments conducted on an open dataset show that the proposed low compute method results in 90.1% mean accuracy on the validation set, which is comparable to the accuracy of a compute-intensive VGG16 network with 94% mean accuracy fine-tuned to this task. To demonstrate the generalizability of our proposed features and model we test the model on BHBID dataset and an internal dataset where we attain accuracy of 98% and 91%, respectively. The proposed method is 10x faster than the VGG16 based model on CPU and scales linearly to the input image size making it suitable to be implemented on low compute edge devices.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 15:39:11 GMT" } ]
2023-04-07T00:00:00
[ [ "Kattamuru", "Sri Charan", "" ], [ "Agrawal", "Kshitij", "" ], [ "Adhikari", "Shyam Prasad", "" ], [ "Bose", "Abhishek", "" ], [ "Misra", "Hemant", "" ] ]
TITLE: Patch-wise Features for Blur Image Classification ABSTRACT: Images captured through smartphone cameras often suffer from degradation, blur being one of the major ones, posing a challenge in processing these images for downstream tasks. In this paper we propose low-compute lightweight patch-wise features for image quality assessment. Using our method we can discriminate between blur vs sharp image degradation. To this end, we train a decision-tree based XGBoost model on various intuitive image features like gray level variance, first and second order gradients, texture features like local binary patterns. Experiments conducted on an open dataset show that the proposed low compute method results in 90.1% mean accuracy on the validation set, which is comparable to the accuracy of a compute-intensive VGG16 network with 94% mean accuracy fine-tuned to this task. To demonstrate the generalizability of our proposed features and model we test the model on BHBID dataset and an internal dataset where we attain accuracy of 98% and 91%, respectively. The proposed method is 10x faster than the VGG16 based model on CPU and scales linearly to the input image size making it suitable to be implemented on low compute edge devices.
no_new_dataset
0.711425
2304.03158
Wu Xing
Xing Wu, Guangyuan Ma, Peng Wang, Meng Lin, Zijia Lin, Fuzheng Zhang and Songlin Hu
CoT-MAE v2: Contextual Masked Auto-Encoder with Multi-view Modeling for Passage Retrieval
working in progress
null
null
null
cs.CL cs.AI cs.IR cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Growing techniques have been emerging to improve the performance of passage retrieval. As an effective representation bottleneck pretraining technique, the contextual masked auto-encoder utilizes contextual embedding to assist in the reconstruction of passages. However, it only uses a single auto-encoding pre-task for dense representation pre-training. This study brings multi-view modeling to the contextual masked auto-encoder. Firstly, multi-view representation utilizes both dense and sparse vectors as multi-view representations, aiming to capture sentence semantics from different aspects. Moreover, multiview decoding paradigm utilizes both autoencoding and auto-regressive decoders in representation bottleneck pre-training, aiming to provide both reconstructive and generative signals for better contextual representation pretraining. We refer to this multi-view pretraining method as CoT-MAE v2. Through extensive experiments, we show that CoT-MAE v2 is effective and robust on large-scale passage retrieval benchmarks and out-of-domain zero-shot benchmarks.
[ { "version": "v1", "created": "Wed, 5 Apr 2023 08:00:38 GMT" } ]
2023-04-07T00:00:00
[ [ "Wu", "Xing", "" ], [ "Ma", "Guangyuan", "" ], [ "Wang", "Peng", "" ], [ "Lin", "Meng", "" ], [ "Lin", "Zijia", "" ], [ "Zhang", "Fuzheng", "" ], [ "Hu", "Songlin", "" ] ]
TITLE: CoT-MAE v2: Contextual Masked Auto-Encoder with Multi-view Modeling for Passage Retrieval ABSTRACT: Growing techniques have been emerging to improve the performance of passage retrieval. As an effective representation bottleneck pretraining technique, the contextual masked auto-encoder utilizes contextual embedding to assist in the reconstruction of passages. However, it only uses a single auto-encoding pre-task for dense representation pre-training. This study brings multi-view modeling to the contextual masked auto-encoder. Firstly, multi-view representation utilizes both dense and sparse vectors as multi-view representations, aiming to capture sentence semantics from different aspects. Moreover, multiview decoding paradigm utilizes both autoencoding and auto-regressive decoders in representation bottleneck pre-training, aiming to provide both reconstructive and generative signals for better contextual representation pretraining. We refer to this multi-view pretraining method as CoT-MAE v2. Through extensive experiments, we show that CoT-MAE v2 is effective and robust on large-scale passage retrieval benchmarks and out-of-domain zero-shot benchmarks.
no_new_dataset
0.707979
2304.03159
Zhichao Duan
Zhichao Duan, Xiuxing Li, Zhengyan Zhang, Zhenyu Li, Ning Liu, Jianyong Wang
Bridging the Language Gap: Knowledge Injected Multilingual Question Answering
ICBK 2021
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
Question Answering (QA) is the task of automatically answering questions posed by humans in natural languages. There are different settings to answer a question, such as abstractive, extractive, boolean, and multiple-choice QA. As a popular topic in natural language processing tasks, extractive question answering task (extractive QA) has gained extensive attention in the past few years. With the continuous evolvement of the world, generalized cross-lingual transfer (G-XLT), where question and answer context are in different languages, poses some unique challenges over cross-lingual transfer (XLT), where question and answer context are in the same language. With the boost of corresponding development of related benchmarks, many works have been done to improve the performance of various language QA tasks. However, only a few works are dedicated to the G-XLT task. In this work, we propose a generalized cross-lingual transfer framework to enhance the model's ability to understand different languages. Specifically, we first assemble triples from different languages to form multilingual knowledge. Since the lack of knowledge between different languages greatly limits models' reasoning ability, we further design a knowledge injection strategy via leveraging link prediction techniques to enrich the model storage of multilingual knowledge. In this way, we can profoundly exploit rich semantic knowledge. Experiment results on real-world datasets MLQA demonstrate that the proposed method can improve the performance by a large margin, outperforming the baseline method by 13.18%/12.00% F1/EM on average.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 15:41:25 GMT" } ]
2023-04-07T00:00:00
[ [ "Duan", "Zhichao", "" ], [ "Li", "Xiuxing", "" ], [ "Zhang", "Zhengyan", "" ], [ "Li", "Zhenyu", "" ], [ "Liu", "Ning", "" ], [ "Wang", "Jianyong", "" ] ]
TITLE: Bridging the Language Gap: Knowledge Injected Multilingual Question Answering ABSTRACT: Question Answering (QA) is the task of automatically answering questions posed by humans in natural languages. There are different settings to answer a question, such as abstractive, extractive, boolean, and multiple-choice QA. As a popular topic in natural language processing tasks, extractive question answering task (extractive QA) has gained extensive attention in the past few years. With the continuous evolvement of the world, generalized cross-lingual transfer (G-XLT), where question and answer context are in different languages, poses some unique challenges over cross-lingual transfer (XLT), where question and answer context are in the same language. With the boost of corresponding development of related benchmarks, many works have been done to improve the performance of various language QA tasks. However, only a few works are dedicated to the G-XLT task. In this work, we propose a generalized cross-lingual transfer framework to enhance the model's ability to understand different languages. Specifically, we first assemble triples from different languages to form multilingual knowledge. Since the lack of knowledge between different languages greatly limits models' reasoning ability, we further design a knowledge injection strategy via leveraging link prediction techniques to enrich the model storage of multilingual knowledge. In this way, we can profoundly exploit rich semantic knowledge. Experiment results on real-world datasets MLQA demonstrate that the proposed method can improve the performance by a large margin, outperforming the baseline method by 13.18%/12.00% F1/EM on average.
no_new_dataset
0.705785
2304.03167
Hongwen Zhang
Hongwen Zhang, Siyou Lin, Ruizhi Shao, Yuxiang Zhang, Zerong Zheng, Han Huang, Yandong Guo, Yebin Liu
CloSET: Modeling Clothed Humans on Continuous Surface with Explicit Template Decomposition
Accepted to CVPR 2023. Project page: https://www.liuyebin.com/closet
null
null
null
cs.CV cs.GR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Creating animatable avatars from static scans requires the modeling of clothing deformations in different poses. Existing learning-based methods typically add pose-dependent deformations upon a minimally-clothed mesh template or a learned implicit template, which have limitations in capturing details or hinder end-to-end learning. In this paper, we revisit point-based solutions and propose to decompose explicit garment-related templates and then add pose-dependent wrinkles to them. In this way, the clothing deformations are disentangled such that the pose-dependent wrinkles can be better learned and applied to unseen poses. Additionally, to tackle the seam artifact issues in recent state-of-the-art point-based methods, we propose to learn point features on a body surface, which establishes a continuous and compact feature space to capture the fine-grained and pose-dependent clothing geometry. To facilitate the research in this field, we also introduce a high-quality scan dataset of humans in real-world clothing. Our approach is validated on two existing datasets and our newly introduced dataset, showing better clothing deformation results in unseen poses. The project page with code and dataset can be found at https://www.liuyebin.com/closet.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 15:50:05 GMT" } ]
2023-04-07T00:00:00
[ [ "Zhang", "Hongwen", "" ], [ "Lin", "Siyou", "" ], [ "Shao", "Ruizhi", "" ], [ "Zhang", "Yuxiang", "" ], [ "Zheng", "Zerong", "" ], [ "Huang", "Han", "" ], [ "Guo", "Yandong", "" ], [ "Liu", "Yebin", "" ] ]
TITLE: CloSET: Modeling Clothed Humans on Continuous Surface with Explicit Template Decomposition ABSTRACT: Creating animatable avatars from static scans requires the modeling of clothing deformations in different poses. Existing learning-based methods typically add pose-dependent deformations upon a minimally-clothed mesh template or a learned implicit template, which have limitations in capturing details or hinder end-to-end learning. In this paper, we revisit point-based solutions and propose to decompose explicit garment-related templates and then add pose-dependent wrinkles to them. In this way, the clothing deformations are disentangled such that the pose-dependent wrinkles can be better learned and applied to unseen poses. Additionally, to tackle the seam artifact issues in recent state-of-the-art point-based methods, we propose to learn point features on a body surface, which establishes a continuous and compact feature space to capture the fine-grained and pose-dependent clothing geometry. To facilitate the research in this field, we also introduce a high-quality scan dataset of humans in real-world clothing. Our approach is validated on two existing datasets and our newly introduced dataset, showing better clothing deformation results in unseen poses. The project page with code and dataset can be found at https://www.liuyebin.com/closet.
new_dataset
0.715436
2304.03195
Xuan Bac Nguyen
Xuan-Bac Nguyen, Chi Nhan Duong, Xin Li, Susan Gauch, Han-Seok Seo, Khoa Luu
Micron-BERT: BERT-based Facial Micro-Expression Recognition
Accepted by CVPR2023
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Micro-expression recognition is one of the most challenging topics in affective computing. It aims to recognize tiny facial movements difficult for humans to perceive in a brief period, i.e., 0.25 to 0.5 seconds. Recent advances in pre-training deep Bidirectional Transformers (BERT) have significantly improved self-supervised learning tasks in computer vision. However, the standard BERT in vision problems is designed to learn only from full images or videos, and the architecture cannot accurately detect details of facial micro-expressions. This paper presents Micron-BERT ($\mu$-BERT), a novel approach to facial micro-expression recognition. The proposed method can automatically capture these movements in an unsupervised manner based on two key ideas. First, we employ Diagonal Micro-Attention (DMA) to detect tiny differences between two frames. Second, we introduce a new Patch of Interest (PoI) module to localize and highlight micro-expression interest regions and simultaneously reduce noisy backgrounds and distractions. By incorporating these components into an end-to-end deep network, the proposed $\mu$-BERT significantly outperforms all previous work in various micro-expression tasks. $\mu$-BERT can be trained on a large-scale unlabeled dataset, i.e., up to 8 million images, and achieves high accuracy on new unseen facial micro-expression datasets. Empirical experiments show $\mu$-BERT consistently outperforms state-of-the-art performance on four micro-expression benchmarks, including SAMM, CASME II, SMIC, and CASME3, by significant margins. Code will be available at \url{https://github.com/uark-cviu/Micron-BERT}
[ { "version": "v1", "created": "Thu, 6 Apr 2023 16:19:09 GMT" } ]
2023-04-07T00:00:00
[ [ "Nguyen", "Xuan-Bac", "" ], [ "Duong", "Chi Nhan", "" ], [ "Li", "Xin", "" ], [ "Gauch", "Susan", "" ], [ "Seo", "Han-Seok", "" ], [ "Luu", "Khoa", "" ] ]
TITLE: Micron-BERT: BERT-based Facial Micro-Expression Recognition ABSTRACT: Micro-expression recognition is one of the most challenging topics in affective computing. It aims to recognize tiny facial movements difficult for humans to perceive in a brief period, i.e., 0.25 to 0.5 seconds. Recent advances in pre-training deep Bidirectional Transformers (BERT) have significantly improved self-supervised learning tasks in computer vision. However, the standard BERT in vision problems is designed to learn only from full images or videos, and the architecture cannot accurately detect details of facial micro-expressions. This paper presents Micron-BERT ($\mu$-BERT), a novel approach to facial micro-expression recognition. The proposed method can automatically capture these movements in an unsupervised manner based on two key ideas. First, we employ Diagonal Micro-Attention (DMA) to detect tiny differences between two frames. Second, we introduce a new Patch of Interest (PoI) module to localize and highlight micro-expression interest regions and simultaneously reduce noisy backgrounds and distractions. By incorporating these components into an end-to-end deep network, the proposed $\mu$-BERT significantly outperforms all previous work in various micro-expression tasks. $\mu$-BERT can be trained on a large-scale unlabeled dataset, i.e., up to 8 million images, and achieves high accuracy on new unseen facial micro-expression datasets. Empirical experiments show $\mu$-BERT consistently outperforms state-of-the-art performance on four micro-expression benchmarks, including SAMM, CASME II, SMIC, and CASME3, by significant margins. Code will be available at \url{https://github.com/uark-cviu/Micron-BERT}
no_new_dataset
0.71027
2304.03199
Sicheng Gao
Bohan Zeng, Xuhui Liu, Sicheng Gao, Boyu Liu, Hong Li, Jianzhuang Liu, Baochang Zhang
Face Animation with an Attribute-Guided Diffusion Model
8 pages, 6 figures
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Face animation has achieved much progress in computer vision. However, prevailing GAN-based methods suffer from unnatural distortions and artifacts due to sophisticated motion deformation. In this paper, we propose a Face Animation framework with an attribute-guided Diffusion Model (FADM), which is the first work to exploit the superior modeling capacity of diffusion models for photo-realistic talking-head generation. To mitigate the uncontrollable synthesis effect of the diffusion model, we design an Attribute-Guided Conditioning Network (AGCN) to adaptively combine the coarse animation features and 3D face reconstruction results, which can incorporate appearance and motion conditions into the diffusion process. These specific designs help FADM rectify unnatural artifacts and distortions, and also enrich high-fidelity facial details through iterative diffusion refinements with accurate animation attributes. FADM can flexibly and effectively improve existing animation videos. Extensive experiments on widely used talking-head benchmarks validate the effectiveness of FADM over prior arts.
[ { "version": "v1", "created": "Thu, 6 Apr 2023 16:22:32 GMT" } ]
2023-04-07T00:00:00
[ [ "Zeng", "Bohan", "" ], [ "Liu", "Xuhui", "" ], [ "Gao", "Sicheng", "" ], [ "Liu", "Boyu", "" ], [ "Li", "Hong", "" ], [ "Liu", "Jianzhuang", "" ], [ "Zhang", "Baochang", "" ] ]
TITLE: Face Animation with an Attribute-Guided Diffusion Model ABSTRACT: Face animation has achieved much progress in computer vision. However, prevailing GAN-based methods suffer from unnatural distortions and artifacts due to sophisticated motion deformation. In this paper, we propose a Face Animation framework with an attribute-guided Diffusion Model (FADM), which is the first work to exploit the superior modeling capacity of diffusion models for photo-realistic talking-head generation. To mitigate the uncontrollable synthesis effect of the diffusion model, we design an Attribute-Guided Conditioning Network (AGCN) to adaptively combine the coarse animation features and 3D face reconstruction results, which can incorporate appearance and motion conditions into the diffusion process. These specific designs help FADM rectify unnatural artifacts and distortions, and also enrich high-fidelity facial details through iterative diffusion refinements with accurate animation attributes. FADM can flexibly and effectively improve existing animation videos. Extensive experiments on widely used talking-head benchmarks validate the effectiveness of FADM over prior arts.
no_new_dataset
0.712782