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
2211.04180
Alexander Ziller
Alexander Ziller, Ayhan Can Erdur, Friederike Jungmann, Daniel Rueckert, Rickmer Braren, Georgios Kaissis
Exploiting segmentation labels and representation learning to forecast therapy response of PDAC patients
null
null
null
null
eess.IV cs.CV
http://creativecommons.org/licenses/by/4.0/
The prediction of pancreatic ductal adenocarcinoma therapy response is a clinically challenging and important task in this high-mortality tumour entity. The training of neural networks able to tackle this challenge is impeded by a lack of large datasets and the difficult anatomical localisation of the pancreas. Here, we propose a hybrid deep neural network pipeline to predict tumour response to initial chemotherapy which is based on the Response Evaluation Criteria in Solid Tumors (RECIST) score, a standardised method for cancer response evaluation by clinicians as well as tumour markers, and clinical evaluation of the patients. We leverage a combination of representation transfer from segmentation to classification, as well as localisation and representation learning. Our approach yields a remarkably data-efficient method able to predict treatment response with a ROC-AUC of 63.7% using only 477 datasets in total.
[ { "version": "v1", "created": "Tue, 8 Nov 2022 11:50:31 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 08:07:01 GMT" } ]
2023-03-31T00:00:00
[ [ "Ziller", "Alexander", "" ], [ "Erdur", "Ayhan Can", "" ], [ "Jungmann", "Friederike", "" ], [ "Rueckert", "Daniel", "" ], [ "Braren", "Rickmer", "" ], [ "Kaissis", "Georgios", "" ] ]
TITLE: Exploiting segmentation labels and representation learning to forecast therapy response of PDAC patients ABSTRACT: The prediction of pancreatic ductal adenocarcinoma therapy response is a clinically challenging and important task in this high-mortality tumour entity. The training of neural networks able to tackle this challenge is impeded by a lack of large datasets and the difficult anatomical localisation of the pancreas. Here, we propose a hybrid deep neural network pipeline to predict tumour response to initial chemotherapy which is based on the Response Evaluation Criteria in Solid Tumors (RECIST) score, a standardised method for cancer response evaluation by clinicians as well as tumour markers, and clinical evaluation of the patients. We leverage a combination of representation transfer from segmentation to classification, as well as localisation and representation learning. Our approach yields a remarkably data-efficient method able to predict treatment response with a ROC-AUC of 63.7% using only 477 datasets in total.
no_new_dataset
0.709868
2211.07717
Nawshad Farruque
Nawshad Farruque, Randy Goebel, Sudhakar Sivapalan and Osmar R. Za\"iane
Deep Temporal Modelling of Clinical Depression through Social Media Text
Tables are properly oriented and some more typos were fixed
null
null
null
cs.CL cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We describe the development of a model to detect user-level clinical depression based on a user's temporal social media posts. Our model uses a Depression Symptoms Detection (DSD) classifier, which is trained on the largest existing samples of clinician annotated tweets for clinical depression symptoms. We subsequently use our DSD model to extract clinically relevant features, e.g., depression scores and their consequent temporal patterns, as well as user posting activity patterns, e.g., quantifying their ``no activity'' or ``silence.'' Furthermore, to evaluate the efficacy of these extracted features, we create three kinds of datasets including a test dataset, from two existing well-known benchmark datasets for user-level depression detection. We then provide accuracy measures based on single features, baseline features and feature ablation tests, at several different levels of temporal granularity. The relevant data distributions and clinical depression detection related settings can be exploited to draw a complete picture of the impact of different features across our created datasets. Finally, we show that, in general, only semantic oriented representation models perform well. However, clinical features may enhance overall performance provided that the training and testing distribution is similar, and there is more data in a user's timeline. The consequence is that the predictive capability of depression scores increase significantly while used in a more sensitive clinical depression detection settings.
[ { "version": "v1", "created": "Fri, 28 Oct 2022 18:31:52 GMT" }, { "version": "v2", "created": "Sat, 25 Mar 2023 06:52:37 GMT" }, { "version": "v3", "created": "Thu, 30 Mar 2023 15:03:17 GMT" } ]
2023-03-31T00:00:00
[ [ "Farruque", "Nawshad", "" ], [ "Goebel", "Randy", "" ], [ "Sivapalan", "Sudhakar", "" ], [ "Zaïane", "Osmar R.", "" ] ]
TITLE: Deep Temporal Modelling of Clinical Depression through Social Media Text ABSTRACT: We describe the development of a model to detect user-level clinical depression based on a user's temporal social media posts. Our model uses a Depression Symptoms Detection (DSD) classifier, which is trained on the largest existing samples of clinician annotated tweets for clinical depression symptoms. We subsequently use our DSD model to extract clinically relevant features, e.g., depression scores and their consequent temporal patterns, as well as user posting activity patterns, e.g., quantifying their ``no activity'' or ``silence.'' Furthermore, to evaluate the efficacy of these extracted features, we create three kinds of datasets including a test dataset, from two existing well-known benchmark datasets for user-level depression detection. We then provide accuracy measures based on single features, baseline features and feature ablation tests, at several different levels of temporal granularity. The relevant data distributions and clinical depression detection related settings can be exploited to draw a complete picture of the impact of different features across our created datasets. Finally, we show that, in general, only semantic oriented representation models perform well. However, clinical features may enhance overall performance provided that the training and testing distribution is similar, and there is more data in a user's timeline. The consequence is that the predictive capability of depression scores increase significantly while used in a more sensitive clinical depression detection settings.
new_dataset
0.712639
2211.09790
James Smith
James Seale Smith, Paola Cascante-Bonilla, Assaf Arbelle, Donghyun Kim, Rameswar Panda, David Cox, Diyi Yang, Zsolt Kira, Rogerio Feris, Leonid Karlinsky
ConStruct-VL: Data-Free Continual Structured VL Concepts Learning
Accepted by the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2023)
null
null
null
cs.LG cs.AI cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recently, large-scale pre-trained Vision-and-Language (VL) foundation models have demonstrated remarkable capabilities in many zero-shot downstream tasks, achieving competitive results for recognizing objects defined by as little as short text prompts. However, it has also been shown that VL models are still brittle in Structured VL Concept (SVLC) reasoning, such as the ability to recognize object attributes, states, and inter-object relations. This leads to reasoning mistakes, which need to be corrected as they occur by teaching VL models the missing SVLC skills; often this must be done using private data where the issue was found, which naturally leads to a data-free continual (no task-id) VL learning setting. In this work, we introduce the first Continual Data-Free Structured VL Concepts Learning (ConStruct-VL) benchmark and show it is challenging for many existing data-free CL strategies. We, therefore, propose a data-free method comprised of a new approach of Adversarial Pseudo-Replay (APR) which generates adversarial reminders of past tasks from past task models. To use this method efficiently, we also propose a continual parameter-efficient Layered-LoRA (LaLo) neural architecture allowing no-memory-cost access to all past models at train time. We show this approach outperforms all data-free methods by as much as ~7% while even matching some levels of experience-replay (prohibitive for applications where data-privacy must be preserved). Our code is publicly available at https://github.com/jamessealesmith/ConStruct-VL
[ { "version": "v1", "created": "Thu, 17 Nov 2022 18:57:03 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 17:59:16 GMT" } ]
2023-03-31T00:00:00
[ [ "Smith", "James Seale", "" ], [ "Cascante-Bonilla", "Paola", "" ], [ "Arbelle", "Assaf", "" ], [ "Kim", "Donghyun", "" ], [ "Panda", "Rameswar", "" ], [ "Cox", "David", "" ], [ "Yang", "Diyi", "" ], [ "Kira", "Zsolt", "" ], [ "Feris", "Rogerio", "" ], [ "Karlinsky", "Leonid", "" ] ]
TITLE: ConStruct-VL: Data-Free Continual Structured VL Concepts Learning ABSTRACT: Recently, large-scale pre-trained Vision-and-Language (VL) foundation models have demonstrated remarkable capabilities in many zero-shot downstream tasks, achieving competitive results for recognizing objects defined by as little as short text prompts. However, it has also been shown that VL models are still brittle in Structured VL Concept (SVLC) reasoning, such as the ability to recognize object attributes, states, and inter-object relations. This leads to reasoning mistakes, which need to be corrected as they occur by teaching VL models the missing SVLC skills; often this must be done using private data where the issue was found, which naturally leads to a data-free continual (no task-id) VL learning setting. In this work, we introduce the first Continual Data-Free Structured VL Concepts Learning (ConStruct-VL) benchmark and show it is challenging for many existing data-free CL strategies. We, therefore, propose a data-free method comprised of a new approach of Adversarial Pseudo-Replay (APR) which generates adversarial reminders of past tasks from past task models. To use this method efficiently, we also propose a continual parameter-efficient Layered-LoRA (LaLo) neural architecture allowing no-memory-cost access to all past models at train time. We show this approach outperforms all data-free methods by as much as ~7% while even matching some levels of experience-replay (prohibitive for applications where data-privacy must be preserved). Our code is publicly available at https://github.com/jamessealesmith/ConStruct-VL
no_new_dataset
0.711638
2211.10598
Chuanfu Shen
Chuanfu Shen, Chao Fan, Wei Wu, Rui Wang, George Q. Huang, Shiqi Yu
LidarGait: Benchmarking 3D Gait Recognition with Point Clouds
15 pages, 15 figures, 4 tables
published on CVPR2023
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Video-based gait recognition has achieved impressive results in constrained scenarios. However, visual cameras neglect human 3D structure information, which limits the feasibility of gait recognition in the 3D wild world. Instead of extracting gait features from images, this work explores precise 3D gait features from point clouds and proposes a simple yet efficient 3D gait recognition framework, termed LidarGait. Our proposed approach projects sparse point clouds into depth maps to learn the representations with 3D geometry information, which outperforms existing point-wise and camera-based methods by a significant margin. Due to the lack of point cloud datasets, we built the first large-scale LiDAR-based gait recognition dataset, SUSTech1K, collected by a LiDAR sensor and an RGB camera. The dataset contains 25,239 sequences from 1,050 subjects and covers many variations, including visibility, views, occlusions, clothing, carrying, and scenes. Extensive experiments show that (1) 3D structure information serves as a significant feature for gait recognition. (2) LidarGait outperforms existing point-based and silhouette-based methods by a significant margin, while it also offers stable cross-view results. (3) The LiDAR sensor is superior to the RGB camera for gait recognition in the outdoor environment. The source code and dataset have been made available at https://lidargait.github.io.
[ { "version": "v1", "created": "Sat, 19 Nov 2022 06:23:08 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 07:51:03 GMT" } ]
2023-03-31T00:00:00
[ [ "Shen", "Chuanfu", "" ], [ "Fan", "Chao", "" ], [ "Wu", "Wei", "" ], [ "Wang", "Rui", "" ], [ "Huang", "George Q.", "" ], [ "Yu", "Shiqi", "" ] ]
TITLE: LidarGait: Benchmarking 3D Gait Recognition with Point Clouds ABSTRACT: Video-based gait recognition has achieved impressive results in constrained scenarios. However, visual cameras neglect human 3D structure information, which limits the feasibility of gait recognition in the 3D wild world. Instead of extracting gait features from images, this work explores precise 3D gait features from point clouds and proposes a simple yet efficient 3D gait recognition framework, termed LidarGait. Our proposed approach projects sparse point clouds into depth maps to learn the representations with 3D geometry information, which outperforms existing point-wise and camera-based methods by a significant margin. Due to the lack of point cloud datasets, we built the first large-scale LiDAR-based gait recognition dataset, SUSTech1K, collected by a LiDAR sensor and an RGB camera. The dataset contains 25,239 sequences from 1,050 subjects and covers many variations, including visibility, views, occlusions, clothing, carrying, and scenes. Extensive experiments show that (1) 3D structure information serves as a significant feature for gait recognition. (2) LidarGait outperforms existing point-based and silhouette-based methods by a significant margin, while it also offers stable cross-view results. (3) The LiDAR sensor is superior to the RGB camera for gait recognition in the outdoor environment. The source code and dataset have been made available at https://lidargait.github.io.
new_dataset
0.714802
2211.12634
Jaehyeok Bae
Jaehyeok Bae, Jae-Han Lee, Seyun Kim
PNI : Industrial Anomaly Detection using Position and Neighborhood Information
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Because anomalous samples cannot be used for training, many anomaly detection and localization methods use pre-trained networks and non-parametric modeling to estimate encoded feature distribution. However, these methods neglect the impact of position and neighborhood information on the distribution of normal features. To overcome this, we propose a new algorithm, \textbf{PNI}, which estimates the normal distribution using conditional probability given neighborhood features, modeled with a multi-layer perceptron network. Moreover, position information is utilized by creating a histogram of representative features at each position. Instead of simply resizing the anomaly map, the proposed method employs an additional refine network trained on synthetic anomaly images to better interpolate and account for the shape and edge of the input image. We conducted experiments on the MVTec AD benchmark dataset and achieved state-of-the-art performance, with \textbf{99.56\%} and \textbf{98.98\%} AUROC scores in anomaly detection and localization, respectively.
[ { "version": "v1", "created": "Tue, 22 Nov 2022 23:45:27 GMT" }, { "version": "v2", "created": "Tue, 29 Nov 2022 02:46:08 GMT" }, { "version": "v3", "created": "Thu, 30 Mar 2023 05:17:19 GMT" } ]
2023-03-31T00:00:00
[ [ "Bae", "Jaehyeok", "" ], [ "Lee", "Jae-Han", "" ], [ "Kim", "Seyun", "" ] ]
TITLE: PNI : Industrial Anomaly Detection using Position and Neighborhood Information ABSTRACT: Because anomalous samples cannot be used for training, many anomaly detection and localization methods use pre-trained networks and non-parametric modeling to estimate encoded feature distribution. However, these methods neglect the impact of position and neighborhood information on the distribution of normal features. To overcome this, we propose a new algorithm, \textbf{PNI}, which estimates the normal distribution using conditional probability given neighborhood features, modeled with a multi-layer perceptron network. Moreover, position information is utilized by creating a histogram of representative features at each position. Instead of simply resizing the anomaly map, the proposed method employs an additional refine network trained on synthetic anomaly images to better interpolate and account for the shape and edge of the input image. We conducted experiments on the MVTec AD benchmark dataset and achieved state-of-the-art performance, with \textbf{99.56\%} and \textbf{98.98\%} AUROC scores in anomaly detection and localization, respectively.
no_new_dataset
0.710829
2211.13218
James Smith
James Seale Smith, Leonid Karlinsky, Vyshnavi Gutta, Paola Cascante-Bonilla, Donghyun Kim, Assaf Arbelle, Rameswar Panda, Rogerio Feris, Zsolt Kira
CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learning
Accepted by the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2023)
null
null
null
cs.CV cs.AI cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Computer vision models suffer from a phenomenon known as catastrophic forgetting when learning novel concepts from continuously shifting training data. Typical solutions for this continual learning problem require extensive rehearsal of previously seen data, which increases memory costs and may violate data privacy. Recently, the emergence of large-scale pre-trained vision transformer models has enabled prompting approaches as an alternative to data-rehearsal. These approaches rely on a key-query mechanism to generate prompts and have been found to be highly resistant to catastrophic forgetting in the well-established rehearsal-free continual learning setting. However, the key mechanism of these methods is not trained end-to-end with the task sequence. Our experiments show that this leads to a reduction in their plasticity, hence sacrificing new task accuracy, and inability to benefit from expanded parameter capacity. We instead propose to learn a set of prompt components which are assembled with input-conditioned weights to produce input-conditioned prompts, resulting in a novel attention-based end-to-end key-query scheme. Our experiments show that we outperform the current SOTA method DualPrompt on established benchmarks by as much as 4.5% in average final accuracy. We also outperform the state of art by as much as 4.4% accuracy on a continual learning benchmark which contains both class-incremental and domain-incremental task shifts, corresponding to many practical settings. Our code is available at https://github.com/GT-RIPL/CODA-Prompt
[ { "version": "v1", "created": "Wed, 23 Nov 2022 18:57:11 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 17:58:59 GMT" } ]
2023-03-31T00:00:00
[ [ "Smith", "James Seale", "" ], [ "Karlinsky", "Leonid", "" ], [ "Gutta", "Vyshnavi", "" ], [ "Cascante-Bonilla", "Paola", "" ], [ "Kim", "Donghyun", "" ], [ "Arbelle", "Assaf", "" ], [ "Panda", "Rameswar", "" ], [ "Feris", "Rogerio", "" ], [ "Kira", "Zsolt", "" ] ]
TITLE: CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learning ABSTRACT: Computer vision models suffer from a phenomenon known as catastrophic forgetting when learning novel concepts from continuously shifting training data. Typical solutions for this continual learning problem require extensive rehearsal of previously seen data, which increases memory costs and may violate data privacy. Recently, the emergence of large-scale pre-trained vision transformer models has enabled prompting approaches as an alternative to data-rehearsal. These approaches rely on a key-query mechanism to generate prompts and have been found to be highly resistant to catastrophic forgetting in the well-established rehearsal-free continual learning setting. However, the key mechanism of these methods is not trained end-to-end with the task sequence. Our experiments show that this leads to a reduction in their plasticity, hence sacrificing new task accuracy, and inability to benefit from expanded parameter capacity. We instead propose to learn a set of prompt components which are assembled with input-conditioned weights to produce input-conditioned prompts, resulting in a novel attention-based end-to-end key-query scheme. Our experiments show that we outperform the current SOTA method DualPrompt on established benchmarks by as much as 4.5% in average final accuracy. We also outperform the state of art by as much as 4.4% accuracy on a continual learning benchmark which contains both class-incremental and domain-incremental task shifts, corresponding to many practical settings. Our code is available at https://github.com/GT-RIPL/CODA-Prompt
no_new_dataset
0.708988
2211.13916
Sangmin Woo
Sangmin Woo, Sumin Lee, Yeonju Park, Muhammad Adi Nugroho, Changick Kim
Towards Good Practices for Missing Modality Robust Action Recognition
AAAI 2023 (Oral); Code: https://github.com/sangminwoo/ActionMAE
null
null
null
cs.CV cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
Standard multi-modal models assume the use of the same modalities in training and inference stages. However, in practice, the environment in which multi-modal models operate may not satisfy such assumption. As such, their performances degrade drastically if any modality is missing in the inference stage. We ask: how can we train a model that is robust to missing modalities? This paper seeks a set of good practices for multi-modal action recognition, with a particular interest in circumstances where some modalities are not available at an inference time. First, we study how to effectively regularize the model during training (e.g., data augmentation). Second, we investigate on fusion methods for robustness to missing modalities: we find that transformer-based fusion shows better robustness for missing modality than summation or concatenation. Third, we propose a simple modular network, ActionMAE, which learns missing modality predictive coding by randomly dropping modality features and tries to reconstruct them with the remaining modality features. Coupling these good practices, we build a model that is not only effective in multi-modal action recognition but also robust to modality missing. Our model achieves the state-of-the-arts on multiple benchmarks and maintains competitive performances even in missing modality scenarios. Codes are available at https://github.com/sangminwoo/ActionMAE.
[ { "version": "v1", "created": "Fri, 25 Nov 2022 06:10:57 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 06:35:24 GMT" } ]
2023-03-31T00:00:00
[ [ "Woo", "Sangmin", "" ], [ "Lee", "Sumin", "" ], [ "Park", "Yeonju", "" ], [ "Nugroho", "Muhammad Adi", "" ], [ "Kim", "Changick", "" ] ]
TITLE: Towards Good Practices for Missing Modality Robust Action Recognition ABSTRACT: Standard multi-modal models assume the use of the same modalities in training and inference stages. However, in practice, the environment in which multi-modal models operate may not satisfy such assumption. As such, their performances degrade drastically if any modality is missing in the inference stage. We ask: how can we train a model that is robust to missing modalities? This paper seeks a set of good practices for multi-modal action recognition, with a particular interest in circumstances where some modalities are not available at an inference time. First, we study how to effectively regularize the model during training (e.g., data augmentation). Second, we investigate on fusion methods for robustness to missing modalities: we find that transformer-based fusion shows better robustness for missing modality than summation or concatenation. Third, we propose a simple modular network, ActionMAE, which learns missing modality predictive coding by randomly dropping modality features and tries to reconstruct them with the remaining modality features. Coupling these good practices, we build a model that is not only effective in multi-modal action recognition but also robust to modality missing. Our model achieves the state-of-the-arts on multiple benchmarks and maintains competitive performances even in missing modality scenarios. Codes are available at https://github.com/sangminwoo/ActionMAE.
no_new_dataset
0.70866
2212.01141
Qianwen Meng
Qianwen Meng, Hangwei Qian, Yong Liu, Lizhen Cui, Yonghui Xu, Zhiqi Shen
MHCCL: Masked Hierarchical Cluster-Wise Contrastive Learning for Multivariate Time Series
accepted by AAAI 2023
null
null
null
cs.LG cs.AI
http://creativecommons.org/licenses/by/4.0/
Learning semantic-rich representations from raw unlabeled time series data is critical for downstream tasks such as classification and forecasting. Contrastive learning has recently shown its promising representation learning capability in the absence of expert annotations. However, existing contrastive approaches generally treat each instance independently, which leads to false negative pairs that share the same semantics. To tackle this problem, we propose MHCCL, a Masked Hierarchical Cluster-wise Contrastive Learning model, which exploits semantic information obtained from the hierarchical structure consisting of multiple latent partitions for multivariate time series. Motivated by the observation that fine-grained clustering preserves higher purity while coarse-grained one reflects higher-level semantics, we propose a novel downward masking strategy to filter out fake negatives and supplement positives by incorporating the multi-granularity information from the clustering hierarchy. In addition, a novel upward masking strategy is designed in MHCCL to remove outliers of clusters at each partition to refine prototypes, which helps speed up the hierarchical clustering process and improves the clustering quality. We conduct experimental evaluations on seven widely-used multivariate time series datasets. The results demonstrate the superiority of MHCCL over the state-of-the-art approaches for unsupervised time series representation learning.
[ { "version": "v1", "created": "Fri, 2 Dec 2022 12:42:53 GMT" }, { "version": "v2", "created": "Tue, 7 Mar 2023 11:39:26 GMT" }, { "version": "v3", "created": "Fri, 24 Mar 2023 03:41:05 GMT" }, { "version": "v4", "created": "Thu, 30 Mar 2023 07:11:56 GMT" } ]
2023-03-31T00:00:00
[ [ "Meng", "Qianwen", "" ], [ "Qian", "Hangwei", "" ], [ "Liu", "Yong", "" ], [ "Cui", "Lizhen", "" ], [ "Xu", "Yonghui", "" ], [ "Shen", "Zhiqi", "" ] ]
TITLE: MHCCL: Masked Hierarchical Cluster-Wise Contrastive Learning for Multivariate Time Series ABSTRACT: Learning semantic-rich representations from raw unlabeled time series data is critical for downstream tasks such as classification and forecasting. Contrastive learning has recently shown its promising representation learning capability in the absence of expert annotations. However, existing contrastive approaches generally treat each instance independently, which leads to false negative pairs that share the same semantics. To tackle this problem, we propose MHCCL, a Masked Hierarchical Cluster-wise Contrastive Learning model, which exploits semantic information obtained from the hierarchical structure consisting of multiple latent partitions for multivariate time series. Motivated by the observation that fine-grained clustering preserves higher purity while coarse-grained one reflects higher-level semantics, we propose a novel downward masking strategy to filter out fake negatives and supplement positives by incorporating the multi-granularity information from the clustering hierarchy. In addition, a novel upward masking strategy is designed in MHCCL to remove outliers of clusters at each partition to refine prototypes, which helps speed up the hierarchical clustering process and improves the clustering quality. We conduct experimental evaluations on seven widely-used multivariate time series datasets. The results demonstrate the superiority of MHCCL over the state-of-the-art approaches for unsupervised time series representation learning.
no_new_dataset
0.710842
2212.03038
Orcun Cetintas
Orcun Cetintas, Guillem Bras\'o, Laura Leal-Taix\'e
Unifying Short and Long-Term Tracking with Graph Hierarchies
CVPR 2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Tracking objects over long videos effectively means solving a spectrum of problems, from short-term association for un-occluded objects to long-term association for objects that are occluded and then reappear in the scene. Methods tackling these two tasks are often disjoint and crafted for specific scenarios, and top-performing approaches are often a mix of techniques, which yields engineering-heavy solutions that lack generality. In this work, we question the need for hybrid approaches and introduce SUSHI, a unified and scalable multi-object tracker. Our approach processes long clips by splitting them into a hierarchy of subclips, which enables high scalability. We leverage graph neural networks to process all levels of the hierarchy, which makes our model unified across temporal scales and highly general. As a result, we obtain significant improvements over state-of-the-art on four diverse datasets. Our code and models are available at bit.ly/sushi-mot.
[ { "version": "v1", "created": "Tue, 6 Dec 2022 15:12:53 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 13:47:25 GMT" } ]
2023-03-31T00:00:00
[ [ "Cetintas", "Orcun", "" ], [ "Brasó", "Guillem", "" ], [ "Leal-Taixé", "Laura", "" ] ]
TITLE: Unifying Short and Long-Term Tracking with Graph Hierarchies ABSTRACT: Tracking objects over long videos effectively means solving a spectrum of problems, from short-term association for un-occluded objects to long-term association for objects that are occluded and then reappear in the scene. Methods tackling these two tasks are often disjoint and crafted for specific scenarios, and top-performing approaches are often a mix of techniques, which yields engineering-heavy solutions that lack generality. In this work, we question the need for hybrid approaches and introduce SUSHI, a unified and scalable multi-object tracker. Our approach processes long clips by splitting them into a hierarchy of subclips, which enables high scalability. We leverage graph neural networks to process all levels of the hierarchy, which makes our model unified across temporal scales and highly general. As a result, we obtain significant improvements over state-of-the-art on four diverse datasets. Our code and models are available at bit.ly/sushi-mot.
no_new_dataset
0.709025
2212.09281
Guang Li
Guang Li, Ren Togo, Takahiro Ogawa, Miki Haseyama
Boosting Automatic COVID-19 Detection Performance with Self-Supervised Learning and Batch Knowledge Ensembling
Published as a journal paper at Elsevier CIBM
null
null
null
eess.IV cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Problem: Detecting COVID-19 from chest X-Ray (CXR) images has become one of the fastest and easiest methods for detecting COVID-19. However, the existing methods usually use supervised transfer learning from natural images as a pretraining process. These methods do not consider the unique features of COVID-19 and the similar features between COVID-19 and other pneumonia. Aim: In this paper, we want to design a novel high-accuracy COVID-19 detection method that uses CXR images, which can consider the unique features of COVID-19 and the similar features between COVID-19 and other pneumonia. Methods: Our method consists of two phases. One is self-supervised learning-based pertaining; the other is batch knowledge ensembling-based fine-tuning. Self-supervised learning-based pretraining can learn distinguished representations from CXR images without manually annotated labels. On the other hand, batch knowledge ensembling-based fine-tuning can utilize category knowledge of images in a batch according to their visual feature similarities to improve detection performance. Unlike our previous implementation, we introduce batch knowledge ensembling into the fine-tuning phase, reducing the memory used in self-supervised learning and improving COVID-19 detection accuracy. Results: On two public COVID-19 CXR datasets, namely, a large dataset and an unbalanced dataset, our method exhibited promising COVID-19 detection performance. Our method maintains high detection accuracy even when annotated CXR training images are reduced significantly (e.g., using only 10% of the original dataset). In addition, our method is insensitive to changes in hyperparameters.
[ { "version": "v1", "created": "Mon, 19 Dec 2022 07:39:26 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 12:32:37 GMT" } ]
2023-03-31T00:00:00
[ [ "Li", "Guang", "" ], [ "Togo", "Ren", "" ], [ "Ogawa", "Takahiro", "" ], [ "Haseyama", "Miki", "" ] ]
TITLE: Boosting Automatic COVID-19 Detection Performance with Self-Supervised Learning and Batch Knowledge Ensembling ABSTRACT: Problem: Detecting COVID-19 from chest X-Ray (CXR) images has become one of the fastest and easiest methods for detecting COVID-19. However, the existing methods usually use supervised transfer learning from natural images as a pretraining process. These methods do not consider the unique features of COVID-19 and the similar features between COVID-19 and other pneumonia. Aim: In this paper, we want to design a novel high-accuracy COVID-19 detection method that uses CXR images, which can consider the unique features of COVID-19 and the similar features between COVID-19 and other pneumonia. Methods: Our method consists of two phases. One is self-supervised learning-based pertaining; the other is batch knowledge ensembling-based fine-tuning. Self-supervised learning-based pretraining can learn distinguished representations from CXR images without manually annotated labels. On the other hand, batch knowledge ensembling-based fine-tuning can utilize category knowledge of images in a batch according to their visual feature similarities to improve detection performance. Unlike our previous implementation, we introduce batch knowledge ensembling into the fine-tuning phase, reducing the memory used in self-supervised learning and improving COVID-19 detection accuracy. Results: On two public COVID-19 CXR datasets, namely, a large dataset and an unbalanced dataset, our method exhibited promising COVID-19 detection performance. Our method maintains high detection accuracy even when annotated CXR training images are reduced significantly (e.g., using only 10% of the original dataset). In addition, our method is insensitive to changes in hyperparameters.
no_new_dataset
0.706981
2301.02280
Filip Radenovic
Filip Radenovic, Abhimanyu Dubey, Abhishek Kadian, Todor Mihaylov, Simon Vandenhende, Yash Patel, Yi Wen, Vignesh Ramanathan, Dhruv Mahajan
Filtering, Distillation, and Hard Negatives for Vision-Language Pre-Training
CVPR 2023
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Vision-language models trained with contrastive learning on large-scale noisy data are becoming increasingly popular for zero-shot recognition problems. In this paper we improve the following three aspects of the contrastive pre-training pipeline: dataset noise, model initialization and the training objective. First, we propose a straightforward filtering strategy titled Complexity, Action, and Text-spotting (CAT) that significantly reduces dataset size, while achieving improved performance across zero-shot vision-language tasks. Next, we propose an approach titled Concept Distillation to leverage strong unimodal representations for contrastive training that does not increase training complexity while outperforming prior work. Finally, we modify the traditional contrastive alignment objective, and propose an importance-sampling approach to up-sample the importance of hard-negatives without adding additional complexity. On an extensive zero-shot benchmark of 29 tasks, our Distilled and Hard-negative Training (DiHT) approach improves on 20 tasks compared to the baseline. Furthermore, for few-shot linear probing, we propose a novel approach that bridges the gap between zero-shot and few-shot performance, substantially improving over prior work. Models are available at https://github.com/facebookresearch/diht.
[ { "version": "v1", "created": "Thu, 5 Jan 2023 19:48:01 GMT" }, { "version": "v2", "created": "Wed, 29 Mar 2023 19:05:14 GMT" } ]
2023-03-31T00:00:00
[ [ "Radenovic", "Filip", "" ], [ "Dubey", "Abhimanyu", "" ], [ "Kadian", "Abhishek", "" ], [ "Mihaylov", "Todor", "" ], [ "Vandenhende", "Simon", "" ], [ "Patel", "Yash", "" ], [ "Wen", "Yi", "" ], [ "Ramanathan", "Vignesh", "" ], [ "Mahajan", "Dhruv", "" ] ]
TITLE: Filtering, Distillation, and Hard Negatives for Vision-Language Pre-Training ABSTRACT: Vision-language models trained with contrastive learning on large-scale noisy data are becoming increasingly popular for zero-shot recognition problems. In this paper we improve the following three aspects of the contrastive pre-training pipeline: dataset noise, model initialization and the training objective. First, we propose a straightforward filtering strategy titled Complexity, Action, and Text-spotting (CAT) that significantly reduces dataset size, while achieving improved performance across zero-shot vision-language tasks. Next, we propose an approach titled Concept Distillation to leverage strong unimodal representations for contrastive training that does not increase training complexity while outperforming prior work. Finally, we modify the traditional contrastive alignment objective, and propose an importance-sampling approach to up-sample the importance of hard-negatives without adding additional complexity. On an extensive zero-shot benchmark of 29 tasks, our Distilled and Hard-negative Training (DiHT) approach improves on 20 tasks compared to the baseline. Furthermore, for few-shot linear probing, we propose a novel approach that bridges the gap between zero-shot and few-shot performance, substantially improving over prior work. Models are available at https://github.com/facebookresearch/diht.
no_new_dataset
0.712988
2301.12700
Xintao Chu
Xintao Chu, Jianping Liu, Jian Wang, Xiaofeng Wang, Yingfei Wang, Meng Wang, Xunxun Gu
CSDR-BERT: a pre-trained scientific dataset match model for Chinese Scientific Dataset Retrieval
null
null
null
null
cs.IR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
As the number of open and shared scientific datasets on the Internet increases under the open science movement, efficiently retrieving these datasets is a crucial task in information retrieval (IR) research. In recent years, the development of large models, particularly the pre-training and fine-tuning paradigm, which involves pre-training on large models and fine-tuning on downstream tasks, has provided new solutions for IR match tasks. In this study, we use the original BERT token in the embedding layer, improve the Sentence-BERT model structure in the model layer by introducing the SimCSE and K-Nearest Neighbors method, and use the cosent loss function in the optimization phase to optimize the target output. Our experimental results show that our model outperforms other competing models on both public and self-built datasets through comparative experiments and ablation implementations. This study explores and validates the feasibility and efficiency of pre-training techniques for semantic retrieval of Chinese scientific datasets.
[ { "version": "v1", "created": "Mon, 30 Jan 2023 07:12:38 GMT" }, { "version": "v2", "created": "Tue, 31 Jan 2023 04:56:10 GMT" }, { "version": "v3", "created": "Thu, 30 Mar 2023 09:18:52 GMT" } ]
2023-03-31T00:00:00
[ [ "Chu", "Xintao", "" ], [ "Liu", "Jianping", "" ], [ "Wang", "Jian", "" ], [ "Wang", "Xiaofeng", "" ], [ "Wang", "Yingfei", "" ], [ "Wang", "Meng", "" ], [ "Gu", "Xunxun", "" ] ]
TITLE: CSDR-BERT: a pre-trained scientific dataset match model for Chinese Scientific Dataset Retrieval ABSTRACT: As the number of open and shared scientific datasets on the Internet increases under the open science movement, efficiently retrieving these datasets is a crucial task in information retrieval (IR) research. In recent years, the development of large models, particularly the pre-training and fine-tuning paradigm, which involves pre-training on large models and fine-tuning on downstream tasks, has provided new solutions for IR match tasks. In this study, we use the original BERT token in the embedding layer, improve the Sentence-BERT model structure in the model layer by introducing the SimCSE and K-Nearest Neighbors method, and use the cosent loss function in the optimization phase to optimize the target output. Our experimental results show that our model outperforms other competing models on both public and self-built datasets through comparative experiments and ablation implementations. This study explores and validates the feasibility and efficiency of pre-training techniques for semantic retrieval of Chinese scientific datasets.
no_new_dataset
0.711782
2302.00388
So Takao
Sean Nassimiha, Peter Dudfield, Jack Kelly, Marc Peter Deisenroth, So Takao
Short-term Prediction and Filtering of Solar Power Using State-Space Gaussian Processes
Workshop paper submitted to "Tackling Climate Change with Machine Learning: workshop at NeurIPS 2022"
null
null
null
cs.LG stat.AP
http://creativecommons.org/licenses/by/4.0/
Short-term forecasting of solar photovoltaic energy (PV) production is important for powerplant management. Ideally these forecasts are equipped with error bars, so that downstream decisions can account for uncertainty. To produce predictions with error bars in this setting, we consider Gaussian processes (GPs) for modelling and predicting solar photovoltaic energy production in the UK. A standard application of GP regression on the PV timeseries data is infeasible due to the large data size and non-Gaussianity of PV readings. However, this is made possible by leveraging recent advances in scalable GP inference, in particular, by using the state-space form of GPs, combined with modern variational inference techniques. The resulting model is not only scalable to large datasets but can also handle continuous data streams via Kalman filtering.
[ { "version": "v1", "created": "Wed, 1 Feb 2023 11:48:10 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 15:12:17 GMT" } ]
2023-03-31T00:00:00
[ [ "Nassimiha", "Sean", "" ], [ "Dudfield", "Peter", "" ], [ "Kelly", "Jack", "" ], [ "Deisenroth", "Marc Peter", "" ], [ "Takao", "So", "" ] ]
TITLE: Short-term Prediction and Filtering of Solar Power Using State-Space Gaussian Processes ABSTRACT: Short-term forecasting of solar photovoltaic energy (PV) production is important for powerplant management. Ideally these forecasts are equipped with error bars, so that downstream decisions can account for uncertainty. To produce predictions with error bars in this setting, we consider Gaussian processes (GPs) for modelling and predicting solar photovoltaic energy production in the UK. A standard application of GP regression on the PV timeseries data is infeasible due to the large data size and non-Gaussianity of PV readings. However, this is made possible by leveraging recent advances in scalable GP inference, in particular, by using the state-space form of GPs, combined with modern variational inference techniques. The resulting model is not only scalable to large datasets but can also handle continuous data streams via Kalman filtering.
no_new_dataset
0.710867
2302.07685
Sihyun Yu
Sihyun Yu, Kihyuk Sohn, Subin Kim, Jinwoo Shin
Video Probabilistic Diffusion Models in Projected Latent Space
CVPR 2023. Project page: https://sihyun.me/PVDM
null
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Despite the remarkable progress in deep generative models, synthesizing high-resolution and temporally coherent videos still remains a challenge due to their high-dimensionality and complex temporal dynamics along with large spatial variations. Recent works on diffusion models have shown their potential to solve this challenge, yet they suffer from severe computation- and memory-inefficiency that limit the scalability. To handle this issue, we propose a novel generative model for videos, coined projected latent video diffusion models (PVDM), a probabilistic diffusion model which learns a video distribution in a low-dimensional latent space and thus can be efficiently trained with high-resolution videos under limited resources. Specifically, PVDM is composed of two components: (a) an autoencoder that projects a given video as 2D-shaped latent vectors that factorize the complex cubic structure of video pixels and (b) a diffusion model architecture specialized for our new factorized latent space and the training/sampling procedure to synthesize videos of arbitrary length with a single model. Experiments on popular video generation datasets demonstrate the superiority of PVDM compared with previous video synthesis methods; e.g., PVDM obtains the FVD score of 639.7 on the UCF-101 long video (128 frames) generation benchmark, which improves 1773.4 of the prior state-of-the-art.
[ { "version": "v1", "created": "Wed, 15 Feb 2023 14:22:34 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 07:08:21 GMT" } ]
2023-03-31T00:00:00
[ [ "Yu", "Sihyun", "" ], [ "Sohn", "Kihyuk", "" ], [ "Kim", "Subin", "" ], [ "Shin", "Jinwoo", "" ] ]
TITLE: Video Probabilistic Diffusion Models in Projected Latent Space ABSTRACT: Despite the remarkable progress in deep generative models, synthesizing high-resolution and temporally coherent videos still remains a challenge due to their high-dimensionality and complex temporal dynamics along with large spatial variations. Recent works on diffusion models have shown their potential to solve this challenge, yet they suffer from severe computation- and memory-inefficiency that limit the scalability. To handle this issue, we propose a novel generative model for videos, coined projected latent video diffusion models (PVDM), a probabilistic diffusion model which learns a video distribution in a low-dimensional latent space and thus can be efficiently trained with high-resolution videos under limited resources. Specifically, PVDM is composed of two components: (a) an autoencoder that projects a given video as 2D-shaped latent vectors that factorize the complex cubic structure of video pixels and (b) a diffusion model architecture specialized for our new factorized latent space and the training/sampling procedure to synthesize videos of arbitrary length with a single model. Experiments on popular video generation datasets demonstrate the superiority of PVDM compared with previous video synthesis methods; e.g., PVDM obtains the FVD score of 639.7 on the UCF-101 long video (128 frames) generation benchmark, which improves 1773.4 of the prior state-of-the-art.
no_new_dataset
0.710069
2302.08646
Tianyue Zheng
Tianyue Zheng, Ang Li, Zhe Chen, Hongbo Wang, and Jun Luo
AutoFed: Heterogeneity-Aware Federated Multimodal Learning for Robust Autonomous Driving
null
null
null
null
cs.LG cs.AI cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Object detection with on-board sensors (e.g., lidar, radar, and camera) play a crucial role in autonomous driving (AD), and these sensors complement each other in modalities. While crowdsensing may potentially exploit these sensors (of huge quantity) to derive more comprehensive knowledge, \textit{federated learning} (FL) appears to be the necessary tool to reach this potential: it enables autonomous vehicles (AVs) to train machine learning models without explicitly sharing raw sensory data. However, the multimodal sensors introduce various data heterogeneity across distributed AVs (e.g., label quantity skews and varied modalities), posing critical challenges to effective FL. To this end, we present AutoFed as a heterogeneity-aware FL framework to fully exploit multimodal sensory data on AVs and thus enable robust AD. Specifically, we first propose a novel model leveraging pseudo-labeling to avoid mistakenly treating unlabeled objects as the background. We also propose an autoencoder-based data imputation method to fill missing data modality (of certain AVs) with the available ones. To further reconcile the heterogeneity, we finally present a client selection mechanism exploiting the similarities among client models to improve both training stability and convergence rate. Our experiments on benchmark dataset confirm that AutoFed substantially improves over status quo approaches in both precision and recall, while demonstrating strong robustness to adverse weather conditions.
[ { "version": "v1", "created": "Fri, 17 Feb 2023 01:31:53 GMT" }, { "version": "v2", "created": "Fri, 24 Feb 2023 12:34:33 GMT" }, { "version": "v3", "created": "Thu, 30 Mar 2023 16:10:25 GMT" } ]
2023-03-31T00:00:00
[ [ "Zheng", "Tianyue", "" ], [ "Li", "Ang", "" ], [ "Chen", "Zhe", "" ], [ "Wang", "Hongbo", "" ], [ "Luo", "Jun", "" ] ]
TITLE: AutoFed: Heterogeneity-Aware Federated Multimodal Learning for Robust Autonomous Driving ABSTRACT: Object detection with on-board sensors (e.g., lidar, radar, and camera) play a crucial role in autonomous driving (AD), and these sensors complement each other in modalities. While crowdsensing may potentially exploit these sensors (of huge quantity) to derive more comprehensive knowledge, \textit{federated learning} (FL) appears to be the necessary tool to reach this potential: it enables autonomous vehicles (AVs) to train machine learning models without explicitly sharing raw sensory data. However, the multimodal sensors introduce various data heterogeneity across distributed AVs (e.g., label quantity skews and varied modalities), posing critical challenges to effective FL. To this end, we present AutoFed as a heterogeneity-aware FL framework to fully exploit multimodal sensory data on AVs and thus enable robust AD. Specifically, we first propose a novel model leveraging pseudo-labeling to avoid mistakenly treating unlabeled objects as the background. We also propose an autoencoder-based data imputation method to fill missing data modality (of certain AVs) with the available ones. To further reconcile the heterogeneity, we finally present a client selection mechanism exploiting the similarities among client models to improve both training stability and convergence rate. Our experiments on benchmark dataset confirm that AutoFed substantially improves over status quo approaches in both precision and recall, while demonstrating strong robustness to adverse weather conditions.
no_new_dataset
0.71145
2302.10574
Weiqin Zhao
Weiqin Zhao, Shujun Wang, Maximus Yeung, Tianye Niu, Lequan Yu
MulGT: Multi-task Graph-Transformer with Task-aware Knowledge Injection and Domain Knowledge-driven Pooling for Whole Slide Image Analysis
AAAI 2023
null
null
null
cs.CV cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Whole slide image (WSI) has been widely used to assist automated diagnosis under the deep learning fields. However, most previous works only discuss the SINGLE task setting which is not aligned with real clinical setting, where pathologists often conduct multiple diagnosis tasks simultaneously. Also, it is commonly recognized that the multi-task learning paradigm can improve learning efficiency by exploiting commonalities and differences across multiple tasks. To this end, we present a novel multi-task framework (i.e., MulGT) for WSI analysis by the specially designed Graph-Transformer equipped with Task-aware Knowledge Injection and Domain Knowledge-driven Graph Pooling modules. Basically, with the Graph Neural Network and Transformer as the building commons, our framework is able to learn task-agnostic low-level local information as well as task-specific high-level global representation. Considering that different tasks in WSI analysis depend on different features and properties, we also design a novel Task-aware Knowledge Injection module to transfer the task-shared graph embedding into task-specific feature spaces to learn more accurate representation for different tasks. Further, we elaborately design a novel Domain Knowledge-driven Graph Pooling module for each task to improve both the accuracy and robustness of different tasks by leveraging different diagnosis patterns of multiple tasks. We evaluated our method on two public WSI datasets from TCGA projects, i.e., esophageal carcinoma and kidney carcinoma. Experimental results show that our method outperforms single-task counterparts and the state-of-theart methods on both tumor typing and staging tasks.
[ { "version": "v1", "created": "Tue, 21 Feb 2023 10:00:58 GMT" }, { "version": "v2", "created": "Tue, 21 Mar 2023 14:10:33 GMT" }, { "version": "v3", "created": "Thu, 30 Mar 2023 08:51:05 GMT" } ]
2023-03-31T00:00:00
[ [ "Zhao", "Weiqin", "" ], [ "Wang", "Shujun", "" ], [ "Yeung", "Maximus", "" ], [ "Niu", "Tianye", "" ], [ "Yu", "Lequan", "" ] ]
TITLE: MulGT: Multi-task Graph-Transformer with Task-aware Knowledge Injection and Domain Knowledge-driven Pooling for Whole Slide Image Analysis ABSTRACT: Whole slide image (WSI) has been widely used to assist automated diagnosis under the deep learning fields. However, most previous works only discuss the SINGLE task setting which is not aligned with real clinical setting, where pathologists often conduct multiple diagnosis tasks simultaneously. Also, it is commonly recognized that the multi-task learning paradigm can improve learning efficiency by exploiting commonalities and differences across multiple tasks. To this end, we present a novel multi-task framework (i.e., MulGT) for WSI analysis by the specially designed Graph-Transformer equipped with Task-aware Knowledge Injection and Domain Knowledge-driven Graph Pooling modules. Basically, with the Graph Neural Network and Transformer as the building commons, our framework is able to learn task-agnostic low-level local information as well as task-specific high-level global representation. Considering that different tasks in WSI analysis depend on different features and properties, we also design a novel Task-aware Knowledge Injection module to transfer the task-shared graph embedding into task-specific feature spaces to learn more accurate representation for different tasks. Further, we elaborately design a novel Domain Knowledge-driven Graph Pooling module for each task to improve both the accuracy and robustness of different tasks by leveraging different diagnosis patterns of multiple tasks. We evaluated our method on two public WSI datasets from TCGA projects, i.e., esophageal carcinoma and kidney carcinoma. Experimental results show that our method outperforms single-task counterparts and the state-of-theart methods on both tumor typing and staging tasks.
no_new_dataset
0.708219
2303.01573
Shubhankar Mangesh Borse
Shubhankar Borse, Debasmit Das, Hyojin Park, Hong Cai, Risheek Garrepalli, Fatih Porikli
DejaVu: Conditional Regenerative Learning to Enhance Dense Prediction
Accepted to CVPR 2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
We present DejaVu, a novel framework which leverages conditional image regeneration as additional supervision during training to improve deep networks for dense prediction tasks such as segmentation, depth estimation, and surface normal prediction. First, we apply redaction to the input image, which removes certain structural information by sparse sampling or selective frequency removal. Next, we use a conditional regenerator, which takes the redacted image and the dense predictions as inputs, and reconstructs the original image by filling in the missing structural information. In the redacted image, structural attributes like boundaries are broken while semantic context is largely preserved. In order to make the regeneration feasible, the conditional generator will then require the structure information from the other input source, i.e., the dense predictions. As such, by including this conditional regeneration objective during training, DejaVu encourages the base network to learn to embed accurate scene structure in its dense prediction. This leads to more accurate predictions with clearer boundaries and better spatial consistency. When it is feasible to leverage additional computation, DejaVu can be extended to incorporate an attention-based regeneration module within the dense prediction network, which further improves accuracy. Through extensive experiments on multiple dense prediction benchmarks such as Cityscapes, COCO, ADE20K, NYUD-v2, and KITTI, we demonstrate the efficacy of employing DejaVu during training, as it outperforms SOTA methods at no added computation cost.
[ { "version": "v1", "created": "Thu, 2 Mar 2023 20:56:36 GMT" }, { "version": "v2", "created": "Wed, 29 Mar 2023 20:02:14 GMT" } ]
2023-03-31T00:00:00
[ [ "Borse", "Shubhankar", "" ], [ "Das", "Debasmit", "" ], [ "Park", "Hyojin", "" ], [ "Cai", "Hong", "" ], [ "Garrepalli", "Risheek", "" ], [ "Porikli", "Fatih", "" ] ]
TITLE: DejaVu: Conditional Regenerative Learning to Enhance Dense Prediction ABSTRACT: We present DejaVu, a novel framework which leverages conditional image regeneration as additional supervision during training to improve deep networks for dense prediction tasks such as segmentation, depth estimation, and surface normal prediction. First, we apply redaction to the input image, which removes certain structural information by sparse sampling or selective frequency removal. Next, we use a conditional regenerator, which takes the redacted image and the dense predictions as inputs, and reconstructs the original image by filling in the missing structural information. In the redacted image, structural attributes like boundaries are broken while semantic context is largely preserved. In order to make the regeneration feasible, the conditional generator will then require the structure information from the other input source, i.e., the dense predictions. As such, by including this conditional regeneration objective during training, DejaVu encourages the base network to learn to embed accurate scene structure in its dense prediction. This leads to more accurate predictions with clearer boundaries and better spatial consistency. When it is feasible to leverage additional computation, DejaVu can be extended to incorporate an attention-based regeneration module within the dense prediction network, which further improves accuracy. Through extensive experiments on multiple dense prediction benchmarks such as Cityscapes, COCO, ADE20K, NYUD-v2, and KITTI, we demonstrate the efficacy of employing DejaVu during training, as it outperforms SOTA methods at no added computation cost.
no_new_dataset
0.710804
2303.05681
Yifei Xin
Yifei Xin, Dongchao Yang, Yuexian Zou
Improving Text-Audio Retrieval by Text-aware Attention Pooling and Prior Matrix Revised Loss
null
null
null
null
cs.SD eess.AS
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In text-audio retrieval (TAR) tasks, due to the heterogeneity of contents between text and audio, the semantic information contained in the text is only similar to certain frames within the audio. Yet, existing works aggregate the entire audio without considering the text, such as mean-pooling over the frames, which is likely to encode misleading audio information not described in the given text. In this paper, we present a text-aware attention pooling (TAP) module for TAR, which is essentially a scaled dot product attention for a text to attend to its most semantically similar frames. Furthermore, previous methods only conduct the softmax for every single-side retrieval, ignoring the potential cross-retrieval information. By exploring the intrinsic prior of each text-audio pair, we introduce a prior matrix revised (PMR) loss to filter the hard case with high (or low) text-to-audio but low (or high) audio-to-text similarity scores, thus achieving the dual optimal match. Experiments show that our TAP significantly outperforms various text-agnostic pooling functions. Moreover, our PMR loss also shows stable performance gains on multiple datasets.
[ { "version": "v1", "created": "Fri, 10 Mar 2023 03:19:38 GMT" }, { "version": "v2", "created": "Tue, 14 Mar 2023 04:48:28 GMT" }, { "version": "v3", "created": "Sun, 19 Mar 2023 11:27:25 GMT" }, { "version": "v4", "created": "Thu, 30 Mar 2023 10:17:02 GMT" } ]
2023-03-31T00:00:00
[ [ "Xin", "Yifei", "" ], [ "Yang", "Dongchao", "" ], [ "Zou", "Yuexian", "" ] ]
TITLE: Improving Text-Audio Retrieval by Text-aware Attention Pooling and Prior Matrix Revised Loss ABSTRACT: In text-audio retrieval (TAR) tasks, due to the heterogeneity of contents between text and audio, the semantic information contained in the text is only similar to certain frames within the audio. Yet, existing works aggregate the entire audio without considering the text, such as mean-pooling over the frames, which is likely to encode misleading audio information not described in the given text. In this paper, we present a text-aware attention pooling (TAP) module for TAR, which is essentially a scaled dot product attention for a text to attend to its most semantically similar frames. Furthermore, previous methods only conduct the softmax for every single-side retrieval, ignoring the potential cross-retrieval information. By exploring the intrinsic prior of each text-audio pair, we introduce a prior matrix revised (PMR) loss to filter the hard case with high (or low) text-to-audio but low (or high) audio-to-text similarity scores, thus achieving the dual optimal match. Experiments show that our TAP significantly outperforms various text-agnostic pooling functions. Moreover, our PMR loss also shows stable performance gains on multiple datasets.
no_new_dataset
0.708395
2303.07337
Qihao Liu
Qihao Liu, Adam Kortylewski, Alan Yuille
PoseExaminer: Automated Testing of Out-of-Distribution Robustness in Human Pose and Shape Estimation
Accepted to CVPR 2023; Code: https://github.com/qihao067/PoseExaminer
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Human pose and shape (HPS) estimation methods achieve remarkable results. However, current HPS benchmarks are mostly designed to test models in scenarios that are similar to the training data. This can lead to critical situations in real-world applications when the observed data differs significantly from the training data and hence is out-of-distribution (OOD). It is therefore important to test and improve the OOD robustness of HPS methods. To address this fundamental problem, we develop a simulator that can be controlled in a fine-grained manner using interpretable parameters to explore the manifold of images of human pose, e.g. by varying poses, shapes, and clothes. We introduce a learning-based testing method, termed PoseExaminer, that automatically diagnoses HPS algorithms by searching over the parameter space of human pose images to find the failure modes. Our strategy for exploring this high-dimensional parameter space is a multi-agent reinforcement learning system, in which the agents collaborate to explore different parts of the parameter space. We show that our PoseExaminer discovers a variety of limitations in current state-of-the-art models that are relevant in real-world scenarios but are missed by current benchmarks. For example, it finds large regions of realistic human poses that are not predicted correctly, as well as reduced performance for humans with skinny and corpulent body shapes. In addition, we show that fine-tuning HPS methods by exploiting the failure modes found by PoseExaminer improve their robustness and even their performance on standard benchmarks by a significant margin. The code are available for research purposes.
[ { "version": "v1", "created": "Mon, 13 Mar 2023 17:58:54 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 04:34:04 GMT" } ]
2023-03-31T00:00:00
[ [ "Liu", "Qihao", "" ], [ "Kortylewski", "Adam", "" ], [ "Yuille", "Alan", "" ] ]
TITLE: PoseExaminer: Automated Testing of Out-of-Distribution Robustness in Human Pose and Shape Estimation ABSTRACT: Human pose and shape (HPS) estimation methods achieve remarkable results. However, current HPS benchmarks are mostly designed to test models in scenarios that are similar to the training data. This can lead to critical situations in real-world applications when the observed data differs significantly from the training data and hence is out-of-distribution (OOD). It is therefore important to test and improve the OOD robustness of HPS methods. To address this fundamental problem, we develop a simulator that can be controlled in a fine-grained manner using interpretable parameters to explore the manifold of images of human pose, e.g. by varying poses, shapes, and clothes. We introduce a learning-based testing method, termed PoseExaminer, that automatically diagnoses HPS algorithms by searching over the parameter space of human pose images to find the failure modes. Our strategy for exploring this high-dimensional parameter space is a multi-agent reinforcement learning system, in which the agents collaborate to explore different parts of the parameter space. We show that our PoseExaminer discovers a variety of limitations in current state-of-the-art models that are relevant in real-world scenarios but are missed by current benchmarks. For example, it finds large regions of realistic human poses that are not predicted correctly, as well as reduced performance for humans with skinny and corpulent body shapes. In addition, we show that fine-tuning HPS methods by exploiting the failure modes found by PoseExaminer improve their robustness and even their performance on standard benchmarks by a significant margin. The code are available for research purposes.
no_new_dataset
0.709069
2303.07489
Junjie Ke
Junjie Ke, Tianhao Zhang, Yilin Wang, Peyman Milanfar, Feng Yang
MRET: Multi-resolution Transformer for Video Quality Assessment
Frontiers Signal Processing in Computational Video and Video Streaming (https://www.frontiersin.org/articles/10.3389/frsip.2023.1137006/full)
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
No-reference video quality assessment (NR-VQA) for user generated content (UGC) is crucial for understanding and improving visual experience. Unlike video recognition tasks, VQA tasks are sensitive to changes in input resolution. Since large amounts of UGC videos nowadays are 720p or above, the fixed and relatively small input used in conventional NR-VQA methods results in missing high-frequency details for many videos. In this paper, we propose a novel Transformer-based NR-VQA framework that preserves the high-resolution quality information. With the multi-resolution input representation and a novel multi-resolution patch sampling mechanism, our method enables a comprehensive view of both the global video composition and local high-resolution details. The proposed approach can effectively aggregate quality information across different granularities in spatial and temporal dimensions, making the model robust to input resolution variations. Our method achieves state-of-the-art performance on large-scale UGC VQA datasets LSVQ and LSVQ-1080p, and on KoNViD-1k and LIVE-VQC without fine-tuning.
[ { "version": "v1", "created": "Mon, 13 Mar 2023 21:48:49 GMT" }, { "version": "v2", "created": "Wed, 29 Mar 2023 18:23:54 GMT" } ]
2023-03-31T00:00:00
[ [ "Ke", "Junjie", "" ], [ "Zhang", "Tianhao", "" ], [ "Wang", "Yilin", "" ], [ "Milanfar", "Peyman", "" ], [ "Yang", "Feng", "" ] ]
TITLE: MRET: Multi-resolution Transformer for Video Quality Assessment ABSTRACT: No-reference video quality assessment (NR-VQA) for user generated content (UGC) is crucial for understanding and improving visual experience. Unlike video recognition tasks, VQA tasks are sensitive to changes in input resolution. Since large amounts of UGC videos nowadays are 720p or above, the fixed and relatively small input used in conventional NR-VQA methods results in missing high-frequency details for many videos. In this paper, we propose a novel Transformer-based NR-VQA framework that preserves the high-resolution quality information. With the multi-resolution input representation and a novel multi-resolution patch sampling mechanism, our method enables a comprehensive view of both the global video composition and local high-resolution details. The proposed approach can effectively aggregate quality information across different granularities in spatial and temporal dimensions, making the model robust to input resolution variations. Our method achieves state-of-the-art performance on large-scale UGC VQA datasets LSVQ and LSVQ-1080p, and on KoNViD-1k and LIVE-VQC without fine-tuning.
no_new_dataset
0.708862
2303.08132
Qihao Liu
Qihao Liu, Junfeng Wu, Yi Jiang, Xiang Bai, Alan Yuille, Song Bai
InstMove: Instance Motion for Object-centric Video Segmentation
Accepted to CVPR 2023; Code: https://github.com/wjf5203/VNext
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Despite significant efforts, cutting-edge video segmentation methods still remain sensitive to occlusion and rapid movement, due to their reliance on the appearance of objects in the form of object embeddings, which are vulnerable to these disturbances. A common solution is to use optical flow to provide motion information, but essentially it only considers pixel-level motion, which still relies on appearance similarity and hence is often inaccurate under occlusion and fast movement. In this work, we study the instance-level motion and present InstMove, which stands for Instance Motion for Object-centric Video Segmentation. In comparison to pixel-wise motion, InstMove mainly relies on instance-level motion information that is free from image feature embeddings, and features physical interpretations, making it more accurate and robust toward occlusion and fast-moving objects. To better fit in with the video segmentation tasks, InstMove uses instance masks to model the physical presence of an object and learns the dynamic model through a memory network to predict its position and shape in the next frame. With only a few lines of code, InstMove can be integrated into current SOTA methods for three different video segmentation tasks and boost their performance. Specifically, we improve the previous arts by 1.5 AP on OVIS dataset, which features heavy occlusions, and 4.9 AP on YouTubeVIS-Long dataset, which mainly contains fast-moving objects. These results suggest that instance-level motion is robust and accurate, and hence serving as a powerful solution in complex scenarios for object-centric video segmentation.
[ { "version": "v1", "created": "Tue, 14 Mar 2023 17:58:44 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 04:23:32 GMT" } ]
2023-03-31T00:00:00
[ [ "Liu", "Qihao", "" ], [ "Wu", "Junfeng", "" ], [ "Jiang", "Yi", "" ], [ "Bai", "Xiang", "" ], [ "Yuille", "Alan", "" ], [ "Bai", "Song", "" ] ]
TITLE: InstMove: Instance Motion for Object-centric Video Segmentation ABSTRACT: Despite significant efforts, cutting-edge video segmentation methods still remain sensitive to occlusion and rapid movement, due to their reliance on the appearance of objects in the form of object embeddings, which are vulnerable to these disturbances. A common solution is to use optical flow to provide motion information, but essentially it only considers pixel-level motion, which still relies on appearance similarity and hence is often inaccurate under occlusion and fast movement. In this work, we study the instance-level motion and present InstMove, which stands for Instance Motion for Object-centric Video Segmentation. In comparison to pixel-wise motion, InstMove mainly relies on instance-level motion information that is free from image feature embeddings, and features physical interpretations, making it more accurate and robust toward occlusion and fast-moving objects. To better fit in with the video segmentation tasks, InstMove uses instance masks to model the physical presence of an object and learns the dynamic model through a memory network to predict its position and shape in the next frame. With only a few lines of code, InstMove can be integrated into current SOTA methods for three different video segmentation tasks and boost their performance. Specifically, we improve the previous arts by 1.5 AP on OVIS dataset, which features heavy occlusions, and 4.9 AP on YouTubeVIS-Long dataset, which mainly contains fast-moving objects. These results suggest that instance-level motion is robust and accurate, and hence serving as a powerful solution in complex scenarios for object-centric video segmentation.
no_new_dataset
0.711982
2303.12670
Wei Li
Wei Li, Jiahao Xie, Chen Change Loy
Correlational Image Modeling for Self-Supervised Visual Pre-Training
Accepted by CVPR 2023
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We introduce Correlational Image Modeling (CIM), a novel and surprisingly effective approach to self-supervised visual pre-training. Our CIM performs a simple pretext task: we randomly crop image regions (exemplars) from an input image (context) and predict correlation maps between the exemplars and the context. Three key designs enable correlational image modeling as a nontrivial and meaningful self-supervisory task. First, to generate useful exemplar-context pairs, we consider cropping image regions with various scales, shapes, rotations, and transformations. Second, we employ a bootstrap learning framework that involves online and target encoders. During pre-training, the former takes exemplars as inputs while the latter converts the context. Third, we model the output correlation maps via a simple cross-attention block, within which the context serves as queries and the exemplars offer values and keys. We show that CIM performs on par or better than the current state of the art on self-supervised and transfer benchmarks.
[ { "version": "v1", "created": "Wed, 22 Mar 2023 15:48:23 GMT" }, { "version": "v2", "created": "Thu, 23 Mar 2023 05:41:37 GMT" }, { "version": "v3", "created": "Thu, 30 Mar 2023 07:55:43 GMT" } ]
2023-03-31T00:00:00
[ [ "Li", "Wei", "" ], [ "Xie", "Jiahao", "" ], [ "Loy", "Chen Change", "" ] ]
TITLE: Correlational Image Modeling for Self-Supervised Visual Pre-Training ABSTRACT: We introduce Correlational Image Modeling (CIM), a novel and surprisingly effective approach to self-supervised visual pre-training. Our CIM performs a simple pretext task: we randomly crop image regions (exemplars) from an input image (context) and predict correlation maps between the exemplars and the context. Three key designs enable correlational image modeling as a nontrivial and meaningful self-supervisory task. First, to generate useful exemplar-context pairs, we consider cropping image regions with various scales, shapes, rotations, and transformations. Second, we employ a bootstrap learning framework that involves online and target encoders. During pre-training, the former takes exemplars as inputs while the latter converts the context. Third, we model the output correlation maps via a simple cross-attention block, within which the context serves as queries and the exemplars offer values and keys. We show that CIM performs on par or better than the current state of the art on self-supervised and transfer benchmarks.
no_new_dataset
0.708591
2303.13750
Qian Tao
Qian Tao, Zhen Wang, Wenyuan Yu, Yaliang Li, Zhewei Wei
LON-GNN: Spectral GNNs with Learnable Orthonormal Basis
null
null
null
null
cs.LG cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In recent years, a plethora of spectral graph neural networks (GNN) methods have utilized polynomial basis with learnable coefficients to achieve top-tier performances on many node-level tasks. Although various kinds of polynomial bases have been explored, each such method adopts a fixed polynomial basis which might not be the optimal choice for the given graph. Besides, we identify the so-called over-passing issue of these methods and show that it is somewhat rooted in their less-principled regularization strategy and unnormalized basis. In this paper, we make the first attempts to address these two issues. Leveraging Jacobi polynomials, we design a novel spectral GNN, LON-GNN, with Learnable OrthoNormal bases and prove that regularizing coefficients becomes equivalent to regularizing the norm of learned filter function now. We conduct extensive experiments on diverse graph datasets to evaluate the fitting and generalization capability of LON-GNN, where the results imply its superiority.
[ { "version": "v1", "created": "Fri, 24 Mar 2023 02:07:46 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 02:25:54 GMT" } ]
2023-03-31T00:00:00
[ [ "Tao", "Qian", "" ], [ "Wang", "Zhen", "" ], [ "Yu", "Wenyuan", "" ], [ "Li", "Yaliang", "" ], [ "Wei", "Zhewei", "" ] ]
TITLE: LON-GNN: Spectral GNNs with Learnable Orthonormal Basis ABSTRACT: In recent years, a plethora of spectral graph neural networks (GNN) methods have utilized polynomial basis with learnable coefficients to achieve top-tier performances on many node-level tasks. Although various kinds of polynomial bases have been explored, each such method adopts a fixed polynomial basis which might not be the optimal choice for the given graph. Besides, we identify the so-called over-passing issue of these methods and show that it is somewhat rooted in their less-principled regularization strategy and unnormalized basis. In this paper, we make the first attempts to address these two issues. Leveraging Jacobi polynomials, we design a novel spectral GNN, LON-GNN, with Learnable OrthoNormal bases and prove that regularizing coefficients becomes equivalent to regularizing the norm of learned filter function now. We conduct extensive experiments on diverse graph datasets to evaluate the fitting and generalization capability of LON-GNN, where the results imply its superiority.
no_new_dataset
0.710239
2303.14080
Paul Hager
Paul Hager, Martin J. Menten, Daniel Rueckert
Best of Both Worlds: Multimodal Contrastive Learning with Tabular and Imaging Data
Accepted in CVPR 2023
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Medical datasets and especially biobanks, often contain extensive tabular data with rich clinical information in addition to images. In practice, clinicians typically have less data, both in terms of diversity and scale, but still wish to deploy deep learning solutions. Combined with increasing medical dataset sizes and expensive annotation costs, the necessity for unsupervised methods that can pretrain multimodally and predict unimodally has risen. To address these needs, we propose the first self-supervised contrastive learning framework that takes advantage of images and tabular data to train unimodal encoders. Our solution combines SimCLR and SCARF, two leading contrastive learning strategies, and is simple and effective. In our experiments, we demonstrate the strength of our framework by predicting risks of myocardial infarction and coronary artery disease (CAD) using cardiac MR images and 120 clinical features from 40,000 UK Biobank subjects. Furthermore, we show the generalizability of our approach to natural images using the DVM car advertisement dataset. We take advantage of the high interpretability of tabular data and through attribution and ablation experiments find that morphometric tabular features, describing size and shape, have outsized importance during the contrastive learning process and improve the quality of the learned embeddings. Finally, we introduce a novel form of supervised contrastive learning, label as a feature (LaaF), by appending the ground truth label as a tabular feature during multimodal pretraining, outperforming all supervised contrastive baselines.
[ { "version": "v1", "created": "Fri, 24 Mar 2023 15:44:42 GMT" }, { "version": "v2", "created": "Mon, 27 Mar 2023 14:24:03 GMT" }, { "version": "v3", "created": "Thu, 30 Mar 2023 12:40:35 GMT" } ]
2023-03-31T00:00:00
[ [ "Hager", "Paul", "" ], [ "Menten", "Martin J.", "" ], [ "Rueckert", "Daniel", "" ] ]
TITLE: Best of Both Worlds: Multimodal Contrastive Learning with Tabular and Imaging Data ABSTRACT: Medical datasets and especially biobanks, often contain extensive tabular data with rich clinical information in addition to images. In practice, clinicians typically have less data, both in terms of diversity and scale, but still wish to deploy deep learning solutions. Combined with increasing medical dataset sizes and expensive annotation costs, the necessity for unsupervised methods that can pretrain multimodally and predict unimodally has risen. To address these needs, we propose the first self-supervised contrastive learning framework that takes advantage of images and tabular data to train unimodal encoders. Our solution combines SimCLR and SCARF, two leading contrastive learning strategies, and is simple and effective. In our experiments, we demonstrate the strength of our framework by predicting risks of myocardial infarction and coronary artery disease (CAD) using cardiac MR images and 120 clinical features from 40,000 UK Biobank subjects. Furthermore, we show the generalizability of our approach to natural images using the DVM car advertisement dataset. We take advantage of the high interpretability of tabular data and through attribution and ablation experiments find that morphometric tabular features, describing size and shape, have outsized importance during the contrastive learning process and improve the quality of the learned embeddings. Finally, we introduce a novel form of supervised contrastive learning, label as a feature (LaaF), by appending the ground truth label as a tabular feature during multimodal pretraining, outperforming all supervised contrastive baselines.
no_new_dataset
0.709025
2303.14522
Jessica M\'egane
Pedro Carvalho and Jessica M\'egane and Nuno Louren\c{c}o and Penousal Machado
Context Matters: Adaptive Mutation for Grammars
16 pages, 6 figures, 5 tables
null
10.1007/978-3-031-29573-7_8
null
cs.NE
http://creativecommons.org/licenses/by/4.0/
This work proposes Adaptive Facilitated Mutation, a self-adaptive mutation method for Structured Grammatical Evolution (SGE), biologically inspired by the theory of facilitated variation. In SGE, the genotype of individuals contains a list for each non-terminal of the grammar that defines the search space. In our proposed mutation, each individual contains an array with a different, self-adaptive mutation rate for each non-terminal. We also propose Function Grouped Grammars, a grammar design procedure, to enhance the benefits of the proposed mutation. Experiments were conducted on three symbolic regression benchmarks using Probabilistic Structured Grammatical Evolution (PSGE), a variant of SGE. Results show our approach is similar or better when compared with the standard grammar and mutation.
[ { "version": "v1", "created": "Sat, 25 Mar 2023 17:26:20 GMT" } ]
2023-03-31T00:00:00
[ [ "Carvalho", "Pedro", "" ], [ "Mégane", "Jessica", "" ], [ "Lourenço", "Nuno", "" ], [ "Machado", "Penousal", "" ] ]
TITLE: Context Matters: Adaptive Mutation for Grammars ABSTRACT: This work proposes Adaptive Facilitated Mutation, a self-adaptive mutation method for Structured Grammatical Evolution (SGE), biologically inspired by the theory of facilitated variation. In SGE, the genotype of individuals contains a list for each non-terminal of the grammar that defines the search space. In our proposed mutation, each individual contains an array with a different, self-adaptive mutation rate for each non-terminal. We also propose Function Grouped Grammars, a grammar design procedure, to enhance the benefits of the proposed mutation. Experiments were conducted on three symbolic regression benchmarks using Probabilistic Structured Grammatical Evolution (PSGE), a variant of SGE. Results show our approach is similar or better when compared with the standard grammar and mutation.
no_new_dataset
0.711857
2303.14956
Xuanfan Ni
Xuanfan Ni, Piji Li and Huayang Li
Unified Text Structuralization with Instruction-tuned Language Models
13 pages, 5 figures
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Text structuralization is one of the important fields of natural language processing (NLP) consists of information extraction (IE) and structure formalization. However, current studies of text structuralization suffer from a shortage of manually annotated high-quality datasets from different domains and languages, which require specialized professional knowledge. In addition, most IE methods are designed for a specific type of structured data, e.g., entities, relations, and events, making them hard to generalize to others. In this work, we propose a simple and efficient approach to instruct large language model (LLM) to extract a variety of structures from texts. More concretely, we add a prefix and a suffix instruction to indicate the desired IE task and structure type, respectively, before feeding the text into a LLM. Experiments on two LLMs show that this approach can enable language models to perform comparable with other state-of-the-art methods on datasets of a variety of languages and knowledge, and can generalize to other IE sub-tasks via changing the content of instruction. Another benefit of our approach is that it can help researchers to build datasets in low-source and domain-specific scenarios, e.g., fields in finance and law, with low cost.
[ { "version": "v1", "created": "Mon, 27 Mar 2023 07:39:05 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 13:41:54 GMT" } ]
2023-03-31T00:00:00
[ [ "Ni", "Xuanfan", "" ], [ "Li", "Piji", "" ], [ "Li", "Huayang", "" ] ]
TITLE: Unified Text Structuralization with Instruction-tuned Language Models ABSTRACT: Text structuralization is one of the important fields of natural language processing (NLP) consists of information extraction (IE) and structure formalization. However, current studies of text structuralization suffer from a shortage of manually annotated high-quality datasets from different domains and languages, which require specialized professional knowledge. In addition, most IE methods are designed for a specific type of structured data, e.g., entities, relations, and events, making them hard to generalize to others. In this work, we propose a simple and efficient approach to instruct large language model (LLM) to extract a variety of structures from texts. More concretely, we add a prefix and a suffix instruction to indicate the desired IE task and structure type, respectively, before feeding the text into a LLM. Experiments on two LLMs show that this approach can enable language models to perform comparable with other state-of-the-art methods on datasets of a variety of languages and knowledge, and can generalize to other IE sub-tasks via changing the content of instruction. Another benefit of our approach is that it can help researchers to build datasets in low-source and domain-specific scenarios, e.g., fields in finance and law, with low cost.
no_new_dataset
0.709019
2303.15466
Han Lin
Han Lin, Guangxing Han, Jiawei Ma, Shiyuan Huang, Xudong Lin, Shih-Fu Chang
Supervised Masked Knowledge Distillation for Few-Shot Transformers
To appear in CVPR 2023
null
null
null
cs.CV cs.AI
http://creativecommons.org/licenses/by/4.0/
Vision Transformers (ViTs) emerge to achieve impressive performance on many data-abundant computer vision tasks by capturing long-range dependencies among local features. However, under few-shot learning (FSL) settings on small datasets with only a few labeled data, ViT tends to overfit and suffers from severe performance degradation due to its absence of CNN-alike inductive bias. Previous works in FSL avoid such problem either through the help of self-supervised auxiliary losses, or through the dextile uses of label information under supervised settings. But the gap between self-supervised and supervised few-shot Transformers is still unfilled. Inspired by recent advances in self-supervised knowledge distillation and masked image modeling (MIM), we propose a novel Supervised Masked Knowledge Distillation model (SMKD) for few-shot Transformers which incorporates label information into self-distillation frameworks. Compared with previous self-supervised methods, we allow intra-class knowledge distillation on both class and patch tokens, and introduce the challenging task of masked patch tokens reconstruction across intra-class images. Experimental results on four few-shot classification benchmark datasets show that our method with simple design outperforms previous methods by a large margin and achieves a new start-of-the-art. Detailed ablation studies confirm the effectiveness of each component of our model. Code for this paper is available here: https://github.com/HL-hanlin/SMKD.
[ { "version": "v1", "created": "Sat, 25 Mar 2023 03:31:46 GMT" }, { "version": "v2", "created": "Wed, 29 Mar 2023 01:59:30 GMT" } ]
2023-03-31T00:00:00
[ [ "Lin", "Han", "" ], [ "Han", "Guangxing", "" ], [ "Ma", "Jiawei", "" ], [ "Huang", "Shiyuan", "" ], [ "Lin", "Xudong", "" ], [ "Chang", "Shih-Fu", "" ] ]
TITLE: Supervised Masked Knowledge Distillation for Few-Shot Transformers ABSTRACT: Vision Transformers (ViTs) emerge to achieve impressive performance on many data-abundant computer vision tasks by capturing long-range dependencies among local features. However, under few-shot learning (FSL) settings on small datasets with only a few labeled data, ViT tends to overfit and suffers from severe performance degradation due to its absence of CNN-alike inductive bias. Previous works in FSL avoid such problem either through the help of self-supervised auxiliary losses, or through the dextile uses of label information under supervised settings. But the gap between self-supervised and supervised few-shot Transformers is still unfilled. Inspired by recent advances in self-supervised knowledge distillation and masked image modeling (MIM), we propose a novel Supervised Masked Knowledge Distillation model (SMKD) for few-shot Transformers which incorporates label information into self-distillation frameworks. Compared with previous self-supervised methods, we allow intra-class knowledge distillation on both class and patch tokens, and introduce the challenging task of masked patch tokens reconstruction across intra-class images. Experimental results on four few-shot classification benchmark datasets show that our method with simple design outperforms previous methods by a large margin and achieves a new start-of-the-art. Detailed ablation studies confirm the effectiveness of each component of our model. Code for this paper is available here: https://github.com/HL-hanlin/SMKD.
no_new_dataset
0.710823
2303.15532
Zhiwei Zhou
Zhiwei Zhou and Erick Elejalde
Stance Inference in Twitter through Graph Convolutional Collaborative Filtering Networks with Minimal Supervision
null
null
null
null
cs.SI
http://creativecommons.org/licenses/by/4.0/
Social Media (SM) has become a stage for people to share thoughts, emotions, opinions, and almost every other aspect of their daily lives. This abundance of human interaction makes SM particularly attractive for social sensing. Especially during polarizing events such as political elections or referendums, users post information and encourage others to support their side, using symbols such as hashtags to represent their attitudes. However, many users choose not to attach hashtags to their messages, use a different language, or show their position only indirectly. Thus, automatically identifying their opinions becomes a more challenging task. To uncover these implicit perspectives, we propose a collaborative filtering model based on Graph Convolutional Networks that exploits the textual content in messages and the rich connections between users and topics. Moreover, our approach only requires a small annotation effort compared to state-of-the-art solutions. Nevertheless, the proposed model achieves competitive performance in predicting individuals' stances. We analyze users' attitudes ahead of two constitutional referendums in Chile in 2020 and 2022. Using two large Twitter datasets, our model achieves improvements of 3.4% in recall and 3.6% in accuracy over the baselines.
[ { "version": "v1", "created": "Mon, 27 Mar 2023 18:14:47 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 16:50:10 GMT" } ]
2023-03-31T00:00:00
[ [ "Zhou", "Zhiwei", "" ], [ "Elejalde", "Erick", "" ] ]
TITLE: Stance Inference in Twitter through Graph Convolutional Collaborative Filtering Networks with Minimal Supervision ABSTRACT: Social Media (SM) has become a stage for people to share thoughts, emotions, opinions, and almost every other aspect of their daily lives. This abundance of human interaction makes SM particularly attractive for social sensing. Especially during polarizing events such as political elections or referendums, users post information and encourage others to support their side, using symbols such as hashtags to represent their attitudes. However, many users choose not to attach hashtags to their messages, use a different language, or show their position only indirectly. Thus, automatically identifying their opinions becomes a more challenging task. To uncover these implicit perspectives, we propose a collaborative filtering model based on Graph Convolutional Networks that exploits the textual content in messages and the rich connections between users and topics. Moreover, our approach only requires a small annotation effort compared to state-of-the-art solutions. Nevertheless, the proposed model achieves competitive performance in predicting individuals' stances. We analyze users' attitudes ahead of two constitutional referendums in Chile in 2020 and 2022. Using two large Twitter datasets, our model achieves improvements of 3.4% in recall and 3.6% in accuracy over the baselines.
no_new_dataset
0.710804
2303.15689
Jiaqi Jin
Jiaqi Jin, Siwei Wang, Zhibin Dong, Xinwang Liu, En Zhu
Deep Incomplete Multi-view Clustering with Cross-view Partial Sample and Prototype Alignment
Accepted by IEEE/CVF Conference on Computer Vision and Pattern Recognition 2023
null
null
null
cs.LG cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The success of existing multi-view clustering relies on the assumption of sample integrity across multiple views. However, in real-world scenarios, samples of multi-view are partially available due to data corruption or sensor failure, which leads to incomplete multi-view clustering study (IMVC). Although several attempts have been proposed to address IMVC, they suffer from the following drawbacks: i) Existing methods mainly adopt cross-view contrastive learning forcing the representations of each sample across views to be exactly the same, which might ignore view discrepancy and flexibility in representations; ii) Due to the absence of non-observed samples across multiple views, the obtained prototypes of clusters might be unaligned and biased, leading to incorrect fusion. To address the above issues, we propose a Cross-view Partial Sample and Prototype Alignment Network (CPSPAN) for Deep Incomplete Multi-view Clustering. Firstly, unlike existing contrastive-based methods, we adopt pair-observed data alignment as 'proxy supervised signals' to guide instance-to-instance correspondence construction among views. Then, regarding of the shifted prototypes in IMVC, we further propose a prototype alignment module to achieve incomplete distribution calibration across views. Extensive experimental results showcase the effectiveness of our proposed modules, attaining noteworthy performance improvements when compared to existing IMVC competitors on benchmark datasets.
[ { "version": "v1", "created": "Tue, 28 Mar 2023 02:31:57 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 13:53:11 GMT" } ]
2023-03-31T00:00:00
[ [ "Jin", "Jiaqi", "" ], [ "Wang", "Siwei", "" ], [ "Dong", "Zhibin", "" ], [ "Liu", "Xinwang", "" ], [ "Zhu", "En", "" ] ]
TITLE: Deep Incomplete Multi-view Clustering with Cross-view Partial Sample and Prototype Alignment ABSTRACT: The success of existing multi-view clustering relies on the assumption of sample integrity across multiple views. However, in real-world scenarios, samples of multi-view are partially available due to data corruption or sensor failure, which leads to incomplete multi-view clustering study (IMVC). Although several attempts have been proposed to address IMVC, they suffer from the following drawbacks: i) Existing methods mainly adopt cross-view contrastive learning forcing the representations of each sample across views to be exactly the same, which might ignore view discrepancy and flexibility in representations; ii) Due to the absence of non-observed samples across multiple views, the obtained prototypes of clusters might be unaligned and biased, leading to incorrect fusion. To address the above issues, we propose a Cross-view Partial Sample and Prototype Alignment Network (CPSPAN) for Deep Incomplete Multi-view Clustering. Firstly, unlike existing contrastive-based methods, we adopt pair-observed data alignment as 'proxy supervised signals' to guide instance-to-instance correspondence construction among views. Then, regarding of the shifted prototypes in IMVC, we further propose a prototype alignment module to achieve incomplete distribution calibration across views. Extensive experimental results showcase the effectiveness of our proposed modules, attaining noteworthy performance improvements when compared to existing IMVC competitors on benchmark datasets.
no_new_dataset
0.709453
2303.16270
Jingwei Sun
Jingwei Sun, Ziyue Xu, Dong Yang, Vishwesh Nath, Wenqi Li, Can Zhao, Daguang Xu, Yiran Chen, Holger R. Roth
Communication-Efficient Vertical Federated Learning with Limited Overlapping Samples
null
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Federated learning is a popular collaborative learning approach that enables clients to train a global model without sharing their local data. Vertical federated learning (VFL) deals with scenarios in which the data on clients have different feature spaces but share some overlapping samples. Existing VFL approaches suffer from high communication costs and cannot deal efficiently with limited overlapping samples commonly seen in the real world. We propose a practical vertical federated learning (VFL) framework called \textbf{one-shot VFL} that can solve the communication bottleneck and the problem of limited overlapping samples simultaneously based on semi-supervised learning. We also propose \textbf{few-shot VFL} to improve the accuracy further with just one more communication round between the server and the clients. In our proposed framework, the clients only need to communicate with the server once or only a few times. We evaluate the proposed VFL framework on both image and tabular datasets. Our methods can improve the accuracy by more than 46.5\% and reduce the communication cost by more than 330$\times$ compared with state-of-the-art VFL methods when evaluated on CIFAR-10. Our code will be made publicly available at \url{https://nvidia.github.io/NVFlare/research/one-shot-vfl}.
[ { "version": "v1", "created": "Tue, 28 Mar 2023 19:30:23 GMT" }, { "version": "v2", "created": "Thu, 30 Mar 2023 00:42:31 GMT" } ]
2023-03-31T00:00:00
[ [ "Sun", "Jingwei", "" ], [ "Xu", "Ziyue", "" ], [ "Yang", "Dong", "" ], [ "Nath", "Vishwesh", "" ], [ "Li", "Wenqi", "" ], [ "Zhao", "Can", "" ], [ "Xu", "Daguang", "" ], [ "Chen", "Yiran", "" ], [ "Roth", "Holger R.", "" ] ]
TITLE: Communication-Efficient Vertical Federated Learning with Limited Overlapping Samples ABSTRACT: Federated learning is a popular collaborative learning approach that enables clients to train a global model without sharing their local data. Vertical federated learning (VFL) deals with scenarios in which the data on clients have different feature spaces but share some overlapping samples. Existing VFL approaches suffer from high communication costs and cannot deal efficiently with limited overlapping samples commonly seen in the real world. We propose a practical vertical federated learning (VFL) framework called \textbf{one-shot VFL} that can solve the communication bottleneck and the problem of limited overlapping samples simultaneously based on semi-supervised learning. We also propose \textbf{few-shot VFL} to improve the accuracy further with just one more communication round between the server and the clients. In our proposed framework, the clients only need to communicate with the server once or only a few times. We evaluate the proposed VFL framework on both image and tabular datasets. Our methods can improve the accuracy by more than 46.5\% and reduce the communication cost by more than 330$\times$ compared with state-of-the-art VFL methods when evaluated on CIFAR-10. Our code will be made publicly available at \url{https://nvidia.github.io/NVFlare/research/one-shot-vfl}.
no_new_dataset
0.710038
2303.16908
Ishaan Kale
Ishaan R Kale, Mandar S Sapre, Ayush Khedkar, Kaustubh Dhamankar, Abhinav Anand, Aayushi Singh
Hybrid ACO-CI Algorithm for Beam Design problems
null
null
null
null
cs.NE cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
A range of complicated real-world problems have inspired the development of several optimization methods. Here, a novel hybrid version of the Ant colony optimization (ACO) method is developed using the sample space reduction technique of the Cohort Intelligence (CI) Algorithm. The algorithm is developed, and accuracy is tested by solving 35 standard benchmark test functions. Furthermore, the constrained version of the algorithm is used to solve two mechanical design problems involving stepped cantilever beams and I-section beams. The effectiveness of the proposed technique of solution is evaluated relative to contemporary algorithmic approaches that are already in use. The results show that our proposed hybrid ACO-CI algorithm will take lesser number of iterations to produce the desired output which means lesser computational time. For the minimization of weight of stepped cantilever beam and deflection in I-section beam a proposed hybrid ACO-CI algorithm yielded best results when compared to other existing algorithms. The proposed work could be investigate for variegated real world applications encompassing domains of engineering, combinatorial and health care problems.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 04:37:14 GMT" } ]
2023-03-31T00:00:00
[ [ "Kale", "Ishaan R", "" ], [ "Sapre", "Mandar S", "" ], [ "Khedkar", "Ayush", "" ], [ "Dhamankar", "Kaustubh", "" ], [ "Anand", "Abhinav", "" ], [ "Singh", "Aayushi", "" ] ]
TITLE: Hybrid ACO-CI Algorithm for Beam Design problems ABSTRACT: A range of complicated real-world problems have inspired the development of several optimization methods. Here, a novel hybrid version of the Ant colony optimization (ACO) method is developed using the sample space reduction technique of the Cohort Intelligence (CI) Algorithm. The algorithm is developed, and accuracy is tested by solving 35 standard benchmark test functions. Furthermore, the constrained version of the algorithm is used to solve two mechanical design problems involving stepped cantilever beams and I-section beams. The effectiveness of the proposed technique of solution is evaluated relative to contemporary algorithmic approaches that are already in use. The results show that our proposed hybrid ACO-CI algorithm will take lesser number of iterations to produce the desired output which means lesser computational time. For the minimization of weight of stepped cantilever beam and deflection in I-section beam a proposed hybrid ACO-CI algorithm yielded best results when compared to other existing algorithms. The proposed work could be investigate for variegated real world applications encompassing domains of engineering, combinatorial and health care problems.
no_new_dataset
0.708824
2303.16914
Carlo Adornetto
Carlo Adornetto and Gianluigi Greco
A New Deep Learning and XAI-Based Algorithm for Features Selection in Genomics
8 pages, 5 figures, Best Doctoral Consortium Paper AIxIA2022 (Udine, Italy)
null
null
null
q-bio.GN cs.AI cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In the field of functional genomics, the analysis of gene expression profiles through Machine and Deep Learning is increasingly providing meaningful insight into a number of diseases. The paper proposes a novel algorithm to perform Feature Selection on genomic-scale data, which exploits the reconstruction capabilities of autoencoders and an ad-hoc defined Explainable Artificial Intelligence-based score in order to select the most informative genes for diagnosis, prognosis, and precision medicine. Results of the application on a Chronic Lymphocytic Leukemia dataset evidence the effectiveness of the algorithm, by identifying and suggesting a set of meaningful genes for further medical investigation.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 16:44:13 GMT" } ]
2023-03-31T00:00:00
[ [ "Adornetto", "Carlo", "" ], [ "Greco", "Gianluigi", "" ] ]
TITLE: A New Deep Learning and XAI-Based Algorithm for Features Selection in Genomics ABSTRACT: In the field of functional genomics, the analysis of gene expression profiles through Machine and Deep Learning is increasingly providing meaningful insight into a number of diseases. The paper proposes a novel algorithm to perform Feature Selection on genomic-scale data, which exploits the reconstruction capabilities of autoencoders and an ad-hoc defined Explainable Artificial Intelligence-based score in order to select the most informative genes for diagnosis, prognosis, and precision medicine. Results of the application on a Chronic Lymphocytic Leukemia dataset evidence the effectiveness of the algorithm, by identifying and suggesting a set of meaningful genes for further medical investigation.
no_new_dataset
0.710434
2303.16938
Vasco Lopes Ferrinho
Vasco Lopes, Bruno Degardin, Lu\'is A. Alexandre
Are Neural Architecture Search Benchmarks Well Designed? A Deeper Look Into Operation Importance
15 pages; 11 figues; 10 tables
null
null
null
cs.LG cs.AI cs.CV stat.ML
http://creativecommons.org/licenses/by/4.0/
Neural Architecture Search (NAS) benchmarks significantly improved the capability of developing and comparing NAS methods while at the same time drastically reduced the computational overhead by providing meta-information about thousands of trained neural networks. However, tabular benchmarks have several drawbacks that can hinder fair comparisons and provide unreliable results. These usually focus on providing a small pool of operations in heavily constrained search spaces -- usually cell-based neural networks with pre-defined outer-skeletons. In this work, we conducted an empirical analysis of the widely used NAS-Bench-101, NAS-Bench-201 and TransNAS-Bench-101 benchmarks in terms of their generability and how different operations influence the performance of the generated architectures. We found that only a subset of the operation pool is required to generate architectures close to the upper-bound of the performance range. Also, the performance distribution is negatively skewed, having a higher density of architectures in the upper-bound range. We consistently found convolution layers to have the highest impact on the architecture's performance, and that specific combination of operations favors top-scoring architectures. These findings shed insights on the correct evaluation and comparison of NAS methods using NAS benchmarks, showing that directly searching on NAS-Bench-201, ImageNet16-120 and TransNAS-Bench-101 produces more reliable results than searching only on CIFAR-10. Furthermore, with this work we provide suggestions for future benchmark evaluations and design. The code used to conduct the evaluations is available at https://github.com/VascoLopes/NAS-Benchmark-Evaluation.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 18:03:28 GMT" } ]
2023-03-31T00:00:00
[ [ "Lopes", "Vasco", "" ], [ "Degardin", "Bruno", "" ], [ "Alexandre", "Luís A.", "" ] ]
TITLE: Are Neural Architecture Search Benchmarks Well Designed? A Deeper Look Into Operation Importance ABSTRACT: Neural Architecture Search (NAS) benchmarks significantly improved the capability of developing and comparing NAS methods while at the same time drastically reduced the computational overhead by providing meta-information about thousands of trained neural networks. However, tabular benchmarks have several drawbacks that can hinder fair comparisons and provide unreliable results. These usually focus on providing a small pool of operations in heavily constrained search spaces -- usually cell-based neural networks with pre-defined outer-skeletons. In this work, we conducted an empirical analysis of the widely used NAS-Bench-101, NAS-Bench-201 and TransNAS-Bench-101 benchmarks in terms of their generability and how different operations influence the performance of the generated architectures. We found that only a subset of the operation pool is required to generate architectures close to the upper-bound of the performance range. Also, the performance distribution is negatively skewed, having a higher density of architectures in the upper-bound range. We consistently found convolution layers to have the highest impact on the architecture's performance, and that specific combination of operations favors top-scoring architectures. These findings shed insights on the correct evaluation and comparison of NAS methods using NAS benchmarks, showing that directly searching on NAS-Bench-201, ImageNet16-120 and TransNAS-Bench-101 produces more reliable results than searching only on CIFAR-10. Furthermore, with this work we provide suggestions for future benchmark evaluations and design. The code used to conduct the evaluations is available at https://github.com/VascoLopes/NAS-Benchmark-Evaluation.
no_new_dataset
0.708843
2303.16947
Tong Zhang
Congpei Qiu, Tong Zhang, Wei Ke, Mathieu Salzmann, Sabine S\"usstrunk
De-coupling and De-positioning Dense Self-supervised Learning
null
null
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Dense Self-Supervised Learning (SSL) methods address the limitations of using image-level feature representations when handling images with multiple objects. Although the dense features extracted by employing segmentation maps and bounding boxes allow networks to perform SSL for each object, we show that they suffer from coupling and positional bias, which arise from the receptive field increasing with layer depth and zero-padding. We address this by introducing three data augmentation strategies, and leveraging them in (i) a decoupling module that aims to robustify the network to variations in the object's surroundings, and (ii) a de-positioning module that encourages the network to discard positional object information. We demonstrate the benefits of our method on COCO and on a new challenging benchmark, OpenImage-MINI, for object classification, semantic segmentation, and object detection. Our extensive experiments evidence the better generalization of our method compared to the SOTA dense SSL methods
[ { "version": "v1", "created": "Wed, 29 Mar 2023 18:07:25 GMT" } ]
2023-03-31T00:00:00
[ [ "Qiu", "Congpei", "" ], [ "Zhang", "Tong", "" ], [ "Ke", "Wei", "" ], [ "Salzmann", "Mathieu", "" ], [ "Süsstrunk", "Sabine", "" ] ]
TITLE: De-coupling and De-positioning Dense Self-supervised Learning ABSTRACT: Dense Self-Supervised Learning (SSL) methods address the limitations of using image-level feature representations when handling images with multiple objects. Although the dense features extracted by employing segmentation maps and bounding boxes allow networks to perform SSL for each object, we show that they suffer from coupling and positional bias, which arise from the receptive field increasing with layer depth and zero-padding. We address this by introducing three data augmentation strategies, and leveraging them in (i) a decoupling module that aims to robustify the network to variations in the object's surroundings, and (ii) a de-positioning module that encourages the network to discard positional object information. We demonstrate the benefits of our method on COCO and on a new challenging benchmark, OpenImage-MINI, for object classification, semantic segmentation, and object detection. Our extensive experiments evidence the better generalization of our method compared to the SOTA dense SSL methods
no_new_dataset
0.638582
2303.16955
Soumyadip Sarkar
Soumyadip Sarkar
A Generative Modeling Approach Using Quantum Gates
null
null
null
null
quant-ph cs.LG
http://creativecommons.org/licenses/by/4.0/
In recent years, quantum computing has emerged as a promising technology for solving complex computational problems. Generative modeling is a technique that allows us to learn and generate new data samples similar to the original dataset. In this paper, we propose a generative modeling approach using quantum gates to generate new samples from a given dataset. We start with a brief introduction to quantum computing and generative modeling. Then, we describe our proposed approach, which involves encoding the dataset into quantum states and using quantum gates to manipulate these states to generate new samples. We also provide mathematical details of our approach and demonstrate its effectiveness through experimental results on various datasets.
[ { "version": "v1", "created": "Sat, 25 Mar 2023 18:01:50 GMT" } ]
2023-03-31T00:00:00
[ [ "Sarkar", "Soumyadip", "" ] ]
TITLE: A Generative Modeling Approach Using Quantum Gates ABSTRACT: In recent years, quantum computing has emerged as a promising technology for solving complex computational problems. Generative modeling is a technique that allows us to learn and generate new data samples similar to the original dataset. In this paper, we propose a generative modeling approach using quantum gates to generate new samples from a given dataset. We start with a brief introduction to quantum computing and generative modeling. Then, we describe our proposed approach, which involves encoding the dataset into quantum states and using quantum gates to manipulate these states to generate new samples. We also provide mathematical details of our approach and demonstrate its effectiveness through experimental results on various datasets.
no_new_dataset
0.713999
2303.16956
Muhammad Akbar Husnoo
Muhammad Akbar Husnoo, Adnan Anwar, Haftu Tasew Reda, Nasser Hosseizadeh, Shama Naz Islam, Abdun Naser Mahmood, Robin Doss
FeDiSa: A Semi-asynchronous Federated Learning Framework for Power System Fault and Cyberattack Discrimination
To appear in IEEE INFOCOM AidTSP 2023
null
null
null
cs.CR cs.DC cs.LG cs.SY eess.SY
http://creativecommons.org/licenses/by/4.0/
With growing security and privacy concerns in the Smart Grid domain, intrusion detection on critical energy infrastructure has become a high priority in recent years. To remedy the challenges of privacy preservation and decentralized power zones with strategic data owners, Federated Learning (FL) has contemporarily surfaced as a viable privacy-preserving alternative which enables collaborative training of attack detection models without requiring the sharing of raw data. To address some of the technical challenges associated with conventional synchronous FL, this paper proposes FeDiSa, a novel Semi-asynchronous Federated learning framework for power system faults and cyberattack Discrimination which takes into account communication latency and stragglers. Specifically, we propose a collaborative training of deep auto-encoder by Supervisory Control and Data Acquisition sub-systems which upload their local model updates to a control centre, which then perform a semi-asynchronous model aggregation for a new global model parameters based on a buffer system and a preset cut-off time. Experiments on the proposed framework using publicly available industrial control systems datasets reveal superior attack detection accuracy whilst preserving data confidentiality and minimizing the adverse effects of communication latency and stragglers. Furthermore, we see a 35% improvement in training time, thus validating the robustness of our proposed method.
[ { "version": "v1", "created": "Tue, 28 Mar 2023 13:34:38 GMT" } ]
2023-03-31T00:00:00
[ [ "Husnoo", "Muhammad Akbar", "" ], [ "Anwar", "Adnan", "" ], [ "Reda", "Haftu Tasew", "" ], [ "Hosseizadeh", "Nasser", "" ], [ "Islam", "Shama Naz", "" ], [ "Mahmood", "Abdun Naser", "" ], [ "Doss", "Robin", "" ] ]
TITLE: FeDiSa: A Semi-asynchronous Federated Learning Framework for Power System Fault and Cyberattack Discrimination ABSTRACT: With growing security and privacy concerns in the Smart Grid domain, intrusion detection on critical energy infrastructure has become a high priority in recent years. To remedy the challenges of privacy preservation and decentralized power zones with strategic data owners, Federated Learning (FL) has contemporarily surfaced as a viable privacy-preserving alternative which enables collaborative training of attack detection models without requiring the sharing of raw data. To address some of the technical challenges associated with conventional synchronous FL, this paper proposes FeDiSa, a novel Semi-asynchronous Federated learning framework for power system faults and cyberattack Discrimination which takes into account communication latency and stragglers. Specifically, we propose a collaborative training of deep auto-encoder by Supervisory Control and Data Acquisition sub-systems which upload their local model updates to a control centre, which then perform a semi-asynchronous model aggregation for a new global model parameters based on a buffer system and a preset cut-off time. Experiments on the proposed framework using publicly available industrial control systems datasets reveal superior attack detection accuracy whilst preserving data confidentiality and minimizing the adverse effects of communication latency and stragglers. Furthermore, we see a 35% improvement in training time, thus validating the robustness of our proposed method.
no_new_dataset
0.709038
2303.16958
Haoran Geng
Haoran Geng, Ziming Li, Yiran Geng, Jiayi Chen, Hao Dong, He Wang
PartManip: Learning Cross-Category Generalizable Part Manipulation Policy from Point Cloud Observations
Accepted by CVPR2023
null
null
null
cs.CV cs.RO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Learning a generalizable object manipulation policy is vital for an embodied agent to work in complex real-world scenes. Parts, as the shared components in different object categories, have the potential to increase the generalization ability of the manipulation policy and achieve cross-category object manipulation. In this work, we build the first large-scale, part-based cross-category object manipulation benchmark, PartManip, which is composed of 11 object categories, 494 objects, and 1432 tasks in 6 task classes. Compared to previous work, our benchmark is also more diverse and realistic, i.e., having more objects and using sparse-view point cloud as input without oracle information like part segmentation. To tackle the difficulties of vision-based policy learning, we first train a state-based expert with our proposed part-based canonicalization and part-aware rewards, and then distill the knowledge to a vision-based student. We also find an expressive backbone is essential to overcome the large diversity of different objects. For cross-category generalization, we introduce domain adversarial learning for domain-invariant feature extraction. Extensive experiments in simulation show that our learned policy can outperform other methods by a large margin, especially on unseen object categories. We also demonstrate our method can successfully manipulate novel objects in the real world.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 18:29:30 GMT" } ]
2023-03-31T00:00:00
[ [ "Geng", "Haoran", "" ], [ "Li", "Ziming", "" ], [ "Geng", "Yiran", "" ], [ "Chen", "Jiayi", "" ], [ "Dong", "Hao", "" ], [ "Wang", "He", "" ] ]
TITLE: PartManip: Learning Cross-Category Generalizable Part Manipulation Policy from Point Cloud Observations ABSTRACT: Learning a generalizable object manipulation policy is vital for an embodied agent to work in complex real-world scenes. Parts, as the shared components in different object categories, have the potential to increase the generalization ability of the manipulation policy and achieve cross-category object manipulation. In this work, we build the first large-scale, part-based cross-category object manipulation benchmark, PartManip, which is composed of 11 object categories, 494 objects, and 1432 tasks in 6 task classes. Compared to previous work, our benchmark is also more diverse and realistic, i.e., having more objects and using sparse-view point cloud as input without oracle information like part segmentation. To tackle the difficulties of vision-based policy learning, we first train a state-based expert with our proposed part-based canonicalization and part-aware rewards, and then distill the knowledge to a vision-based student. We also find an expressive backbone is essential to overcome the large diversity of different objects. For cross-category generalization, we introduce domain adversarial learning for domain-invariant feature extraction. Extensive experiments in simulation show that our learned policy can outperform other methods by a large margin, especially on unseen object categories. We also demonstrate our method can successfully manipulate novel objects in the real world.
no_new_dataset
0.696552
2303.16974
Mitchell DeHaven
Mitchell DeHaven and Stephen Scott
BEVERS: A General, Simple, and Performant Framework for Automatic Fact Verification
Accepted to the Sixth FEVER Workshop at EACL 2023
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
Automatic fact verification has become an increasingly popular topic in recent years and among datasets the Fact Extraction and VERification (FEVER) dataset is one of the most popular. In this work we present BEVERS, a tuned baseline system for the FEVER dataset. Our pipeline uses standard approaches for document retrieval, sentence selection, and final claim classification, however, we spend considerable effort ensuring optimal performance for each component. The results are that BEVERS achieves the highest FEVER score and label accuracy among all systems, published or unpublished. We also apply this pipeline to another fact verification dataset, Scifact, and achieve the highest label accuracy among all systems on that dataset as well. We also make our full code available.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 19:16:19 GMT" } ]
2023-03-31T00:00:00
[ [ "DeHaven", "Mitchell", "" ], [ "Scott", "Stephen", "" ] ]
TITLE: BEVERS: A General, Simple, and Performant Framework for Automatic Fact Verification ABSTRACT: Automatic fact verification has become an increasingly popular topic in recent years and among datasets the Fact Extraction and VERification (FEVER) dataset is one of the most popular. In this work we present BEVERS, a tuned baseline system for the FEVER dataset. Our pipeline uses standard approaches for document retrieval, sentence selection, and final claim classification, however, we spend considerable effort ensuring optimal performance for each component. The results are that BEVERS achieves the highest FEVER score and label accuracy among all systems, published or unpublished. We also apply this pipeline to another fact verification dataset, Scifact, and achieve the highest label accuracy among all systems on that dataset as well. We also make our full code available.
no_new_dataset
0.70695
2303.17006
Saranya Venkatraman
Saranya Venkatraman, He He and David Reitter
How do decoding algorithms distribute information in dialogue responses?
null
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
Humans tend to follow the Uniform Information Density (UID) principle by distributing information evenly in utterances. We study if decoding algorithms implicitly follow this UID principle, and under what conditions adherence to UID might be desirable for dialogue generation. We generate responses using different decoding algorithms with GPT-2 on the Persona-Chat dataset and collect human judgments on their quality using Amazon Mechanical Turk. We find that (i) surprisingly, model-generated responses follow the UID principle to a greater extent than human responses, and (ii) decoding algorithms that promote UID do not generate higher-quality responses. Instead, when we control for surprisal, non-uniformity of information density correlates with the quality of responses with very low/high surprisal. Our findings indicate that encouraging non-uniform responses is a potential solution to the ``likelihood trap'' problem (quality degradation in very high-likelihood text). Our dataset containing multiple candidate responses per dialog history along with human-annotated quality ratings is available at https://huggingface.co/datasets/saranya132/dialog_uid_gpt2.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 20:21:45 GMT" } ]
2023-03-31T00:00:00
[ [ "Venkatraman", "Saranya", "" ], [ "He", "He", "" ], [ "Reitter", "David", "" ] ]
TITLE: How do decoding algorithms distribute information in dialogue responses? ABSTRACT: Humans tend to follow the Uniform Information Density (UID) principle by distributing information evenly in utterances. We study if decoding algorithms implicitly follow this UID principle, and under what conditions adherence to UID might be desirable for dialogue generation. We generate responses using different decoding algorithms with GPT-2 on the Persona-Chat dataset and collect human judgments on their quality using Amazon Mechanical Turk. We find that (i) surprisingly, model-generated responses follow the UID principle to a greater extent than human responses, and (ii) decoding algorithms that promote UID do not generate higher-quality responses. Instead, when we control for surprisal, non-uniformity of information density correlates with the quality of responses with very low/high surprisal. Our findings indicate that encouraging non-uniform responses is a potential solution to the ``likelihood trap'' problem (quality degradation in very high-likelihood text). Our dataset containing multiple candidate responses per dialog history along with human-annotated quality ratings is available at https://huggingface.co/datasets/saranya132/dialog_uid_gpt2.
new_dataset
0.71607
2303.17027
Zihao Sheng
Zihao Sheng, Zilin Huang, Sikai Chen
EPG-MGCN: Ego-Planning Guided Multi-Graph Convolutional Network for Heterogeneous Agent Trajectory Prediction
null
null
null
null
cs.LG cs.CV cs.RO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
To drive safely in complex traffic environments, autonomous vehicles need to make an accurate prediction of the future trajectories of nearby heterogeneous traffic agents (i.e., vehicles, pedestrians, bicyclists, etc). Due to the interactive nature, human drivers are accustomed to infer what the future situations will become if they are going to execute different maneuvers. To fully exploit the impacts of interactions, this paper proposes a ego-planning guided multi-graph convolutional network (EPG-MGCN) to predict the trajectories of heterogeneous agents using both historical trajectory information and ego vehicle's future planning information. The EPG-MGCN first models the social interactions by employing four graph topologies, i.e., distance graphs, visibility graphs, planning graphs and category graphs. Then, the planning information of the ego vehicle is encoded by both the planning graph and the subsequent planning-guided prediction module to reduce uncertainty in the trajectory prediction. Finally, a category-specific gated recurrent unit (CS-GRU) encoder-decoder is designed to generate future trajectories for each specific type of agents. Our network is evaluated on two real-world trajectory datasets: ApolloScape and NGSIM. The experimental results show that the proposed EPG-MGCN achieves state-of-the-art performance compared to existing methods.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 21:14:05 GMT" } ]
2023-03-31T00:00:00
[ [ "Sheng", "Zihao", "" ], [ "Huang", "Zilin", "" ], [ "Chen", "Sikai", "" ] ]
TITLE: EPG-MGCN: Ego-Planning Guided Multi-Graph Convolutional Network for Heterogeneous Agent Trajectory Prediction ABSTRACT: To drive safely in complex traffic environments, autonomous vehicles need to make an accurate prediction of the future trajectories of nearby heterogeneous traffic agents (i.e., vehicles, pedestrians, bicyclists, etc). Due to the interactive nature, human drivers are accustomed to infer what the future situations will become if they are going to execute different maneuvers. To fully exploit the impacts of interactions, this paper proposes a ego-planning guided multi-graph convolutional network (EPG-MGCN) to predict the trajectories of heterogeneous agents using both historical trajectory information and ego vehicle's future planning information. The EPG-MGCN first models the social interactions by employing four graph topologies, i.e., distance graphs, visibility graphs, planning graphs and category graphs. Then, the planning information of the ego vehicle is encoded by both the planning graph and the subsequent planning-guided prediction module to reduce uncertainty in the trajectory prediction. Finally, a category-specific gated recurrent unit (CS-GRU) encoder-decoder is designed to generate future trajectories for each specific type of agents. Our network is evaluated on two real-world trajectory datasets: ApolloScape and NGSIM. The experimental results show that the proposed EPG-MGCN achieves state-of-the-art performance compared to existing methods.
no_new_dataset
0.709069
2303.17056
Shentong Mo
Shentong Mo, Yapeng Tian
Audio-Visual Grouping Network for Sound Localization from Mixtures
CVPR 2023
null
null
null
cs.CV cs.LG cs.MM
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sound source localization is a typical and challenging task that predicts the location of sound sources in a video. Previous single-source methods mainly used the audio-visual association as clues to localize sounding objects in each image. Due to the mixed property of multiple sound sources in the original space, there exist rare multi-source approaches to localizing multiple sources simultaneously, except for one recent work using a contrastive random walk in the graph with images and separated sound as nodes. Despite their promising performance, they can only handle a fixed number of sources, and they cannot learn compact class-aware representations for individual sources. To alleviate this shortcoming, in this paper, we propose a novel audio-visual grouping network, namely AVGN, that can directly learn category-wise semantic features for each source from the input audio mixture and image to localize multiple sources simultaneously. Specifically, our AVGN leverages learnable audio-visual class tokens to aggregate class-aware source features. Then, the aggregated semantic features for each source can be used as guidance to localize the corresponding visual regions. Compared to existing multi-source methods, our new framework can localize a flexible number of sources and disentangle category-aware audio-visual representations for individual sound sources. We conduct extensive experiments on MUSIC, VGGSound-Instruments, and VGG-Sound Sources benchmarks. The results demonstrate that the proposed AVGN can achieve state-of-the-art sounding object localization performance on both single-source and multi-source scenarios. Code is available at \url{https://github.com/stoneMo/AVGN}.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 22:58:55 GMT" } ]
2023-03-31T00:00:00
[ [ "Mo", "Shentong", "" ], [ "Tian", "Yapeng", "" ] ]
TITLE: Audio-Visual Grouping Network for Sound Localization from Mixtures ABSTRACT: Sound source localization is a typical and challenging task that predicts the location of sound sources in a video. Previous single-source methods mainly used the audio-visual association as clues to localize sounding objects in each image. Due to the mixed property of multiple sound sources in the original space, there exist rare multi-source approaches to localizing multiple sources simultaneously, except for one recent work using a contrastive random walk in the graph with images and separated sound as nodes. Despite their promising performance, they can only handle a fixed number of sources, and they cannot learn compact class-aware representations for individual sources. To alleviate this shortcoming, in this paper, we propose a novel audio-visual grouping network, namely AVGN, that can directly learn category-wise semantic features for each source from the input audio mixture and image to localize multiple sources simultaneously. Specifically, our AVGN leverages learnable audio-visual class tokens to aggregate class-aware source features. Then, the aggregated semantic features for each source can be used as guidance to localize the corresponding visual regions. Compared to existing multi-source methods, our new framework can localize a flexible number of sources and disentangle category-aware audio-visual representations for individual sound sources. We conduct extensive experiments on MUSIC, VGGSound-Instruments, and VGG-Sound Sources benchmarks. The results demonstrate that the proposed AVGN can achieve state-of-the-art sounding object localization performance on both single-source and multi-source scenarios. Code is available at \url{https://github.com/stoneMo/AVGN}.
no_new_dataset
0.711406
2303.17071
Elliot Schumacher
Varun Nair, Elliot Schumacher, Geoffrey Tso, Anitha Kannan
DERA: Enhancing Large Language Model Completions with Dialog-Enabled Resolving Agents
null
null
null
null
cs.CL cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
Large language models (LLMs) have emerged as valuable tools for many natural language understanding tasks. In safety-critical applications such as healthcare, the utility of these models is governed by their ability to generate outputs that are factually accurate and complete. In this work, we present dialog-enabled resolving agents (DERA). DERA is a paradigm made possible by the increased conversational abilities of LLMs, namely GPT-4. It provides a simple, interpretable forum for models to communicate feedback and iteratively improve output. We frame our dialog as a discussion between two agent types - a Researcher, who processes information and identifies crucial problem components, and a Decider, who has the autonomy to integrate the Researcher's information and makes judgments on the final output. We test DERA against three clinically-focused tasks. For medical conversation summarization and care plan generation, DERA shows significant improvement over the base GPT-4 performance in both human expert preference evaluations and quantitative metrics. In a new finding, we also show that GPT-4's performance (70%) on an open-ended version of the MedQA question-answering (QA) dataset (Jin et al. 2021, USMLE) is well above the passing level (60%), with DERA showing similar performance. We release the open-ended MEDQA dataset at https://github.com/curai/curai-research/tree/main/DERA.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 00:30:19 GMT" } ]
2023-03-31T00:00:00
[ [ "Nair", "Varun", "" ], [ "Schumacher", "Elliot", "" ], [ "Tso", "Geoffrey", "" ], [ "Kannan", "Anitha", "" ] ]
TITLE: DERA: Enhancing Large Language Model Completions with Dialog-Enabled Resolving Agents ABSTRACT: Large language models (LLMs) have emerged as valuable tools for many natural language understanding tasks. In safety-critical applications such as healthcare, the utility of these models is governed by their ability to generate outputs that are factually accurate and complete. In this work, we present dialog-enabled resolving agents (DERA). DERA is a paradigm made possible by the increased conversational abilities of LLMs, namely GPT-4. It provides a simple, interpretable forum for models to communicate feedback and iteratively improve output. We frame our dialog as a discussion between two agent types - a Researcher, who processes information and identifies crucial problem components, and a Decider, who has the autonomy to integrate the Researcher's information and makes judgments on the final output. We test DERA against three clinically-focused tasks. For medical conversation summarization and care plan generation, DERA shows significant improvement over the base GPT-4 performance in both human expert preference evaluations and quantitative metrics. In a new finding, we also show that GPT-4's performance (70%) on an open-ended version of the MedQA question-answering (QA) dataset (Jin et al. 2021, USMLE) is well above the passing level (60%), with DERA showing similar performance. We release the open-ended MEDQA dataset at https://github.com/curai/curai-research/tree/main/DERA.
no_new_dataset
0.683406
2303.17080
Ethan Wisdom
Ethan Wisdom, Tejas Gokhale, Chaowei Xiao, Yezhou Yang
Mole Recruitment: Poisoning of Image Classifiers via Selective Batch Sampling
null
null
null
null
cs.LG
http://creativecommons.org/licenses/by-nc-nd/4.0/
In this work, we present a data poisoning attack that confounds machine learning models without any manipulation of the image or label. This is achieved by simply leveraging the most confounding natural samples found within the training data itself, in a new form of a targeted attack coined "Mole Recruitment." We define moles as the training samples of a class that appear most similar to samples of another class, and show that simply restructuring training batches with an optimal number of moles can lead to significant degradation in the performance of the targeted class. We show the efficacy of this novel attack in an offline setting across several standard image classification datasets, and demonstrate the real-world viability of this attack in a continual learning (CL) setting. Our analysis reveals that state-of-the-art models are susceptible to Mole Recruitment, thereby exposing a previously undetected vulnerability of image classifiers.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 00:59:37 GMT" } ]
2023-03-31T00:00:00
[ [ "Wisdom", "Ethan", "" ], [ "Gokhale", "Tejas", "" ], [ "Xiao", "Chaowei", "" ], [ "Yang", "Yezhou", "" ] ]
TITLE: Mole Recruitment: Poisoning of Image Classifiers via Selective Batch Sampling ABSTRACT: In this work, we present a data poisoning attack that confounds machine learning models without any manipulation of the image or label. This is achieved by simply leveraging the most confounding natural samples found within the training data itself, in a new form of a targeted attack coined "Mole Recruitment." We define moles as the training samples of a class that appear most similar to samples of another class, and show that simply restructuring training batches with an optimal number of moles can lead to significant degradation in the performance of the targeted class. We show the efficacy of this novel attack in an offline setting across several standard image classification datasets, and demonstrate the real-world viability of this attack in a continual learning (CL) setting. Our analysis reveals that state-of-the-art models are susceptible to Mole Recruitment, thereby exposing a previously undetected vulnerability of image classifiers.
no_new_dataset
0.711813
2303.17094
Dr. Suryansh Kumar
Nishant Jain, Suryansh Kumar, Luc Van Gool
Enhanced Stable View Synthesis
Accepted to IEEE/CVF CVPR 2023. Draft info: 13 pages, 6 Figures, 7 Tables
null
null
null
cs.CV cs.GR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We introduce an approach to enhance the novel view synthesis from images taken from a freely moving camera. The introduced approach focuses on outdoor scenes where recovering accurate geometric scaffold and camera pose is challenging, leading to inferior results using the state-of-the-art stable view synthesis (SVS) method. SVS and related methods fail for outdoor scenes primarily due to (i) over-relying on the multiview stereo (MVS) for geometric scaffold recovery and (ii) assuming COLMAP computed camera poses as the best possible estimates, despite it being well-studied that MVS 3D reconstruction accuracy is limited to scene disparity and camera-pose accuracy is sensitive to key-point correspondence selection. This work proposes a principled way to enhance novel view synthesis solutions drawing inspiration from the basics of multiple view geometry. By leveraging the complementary behavior of MVS and monocular depth, we arrive at a better scene depth per view for nearby and far points, respectively. Moreover, our approach jointly refines camera poses with image-based rendering via multiple rotation averaging graph optimization. The recovered scene depth and the camera-pose help better view-dependent on-surface feature aggregation of the entire scene. Extensive evaluation of our approach on the popular benchmark dataset, such as Tanks and Temples, shows substantial improvement in view synthesis results compared to the prior art. For instance, our method shows 1.5 dB of PSNR improvement on the Tank and Temples. Similar statistics are observed when tested on other benchmark datasets such as FVS, Mip-NeRF 360, and DTU.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 01:53:14 GMT" } ]
2023-03-31T00:00:00
[ [ "Jain", "Nishant", "" ], [ "Kumar", "Suryansh", "" ], [ "Van Gool", "Luc", "" ] ]
TITLE: Enhanced Stable View Synthesis ABSTRACT: We introduce an approach to enhance the novel view synthesis from images taken from a freely moving camera. The introduced approach focuses on outdoor scenes where recovering accurate geometric scaffold and camera pose is challenging, leading to inferior results using the state-of-the-art stable view synthesis (SVS) method. SVS and related methods fail for outdoor scenes primarily due to (i) over-relying on the multiview stereo (MVS) for geometric scaffold recovery and (ii) assuming COLMAP computed camera poses as the best possible estimates, despite it being well-studied that MVS 3D reconstruction accuracy is limited to scene disparity and camera-pose accuracy is sensitive to key-point correspondence selection. This work proposes a principled way to enhance novel view synthesis solutions drawing inspiration from the basics of multiple view geometry. By leveraging the complementary behavior of MVS and monocular depth, we arrive at a better scene depth per view for nearby and far points, respectively. Moreover, our approach jointly refines camera poses with image-based rendering via multiple rotation averaging graph optimization. The recovered scene depth and the camera-pose help better view-dependent on-surface feature aggregation of the entire scene. Extensive evaluation of our approach on the popular benchmark dataset, such as Tanks and Temples, shows substantial improvement in view synthesis results compared to the prior art. For instance, our method shows 1.5 dB of PSNR improvement on the Tank and Temples. Similar statistics are observed when tested on other benchmark datasets such as FVS, Mip-NeRF 360, and DTU.
no_new_dataset
0.712589
2303.17096
Xiaodan Li
Xiaodan Li, Yuefeng Chen, Yao Zhu, Shuhui Wang, Rong Zhang, Hui Xue
ImageNet-E: Benchmarking Neural Network Robustness via Attribute Editing
Accepted by CVPR2023
CVPR 2023
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Recent studies have shown that higher accuracy on ImageNet usually leads to better robustness against different corruptions. Therefore, in this paper, instead of following the traditional research paradigm that investigates new out-of-distribution corruptions or perturbations deep models may encounter, we conduct model debugging in in-distribution data to explore which object attributes a model may be sensitive to. To achieve this goal, we create a toolkit for object editing with controls of backgrounds, sizes, positions, and directions, and create a rigorous benchmark named ImageNet-E(diting) for evaluating the image classifier robustness in terms of object attributes. With our ImageNet-E, we evaluate the performance of current deep learning models, including both convolutional neural networks and vision transformers. We find that most models are quite sensitive to attribute changes. A small change in the background can lead to an average of 9.23\% drop on top-1 accuracy. We also evaluate some robust models including both adversarially trained models and other robust trained models and find that some models show worse robustness against attribute changes than vanilla models. Based on these findings, we discover ways to enhance attribute robustness with preprocessing, architecture designs, and training strategies. We hope this work can provide some insights to the community and open up a new avenue for research in robust computer vision. The code and dataset are available at https://github.com/alibaba/easyrobust.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 02:02:32 GMT" } ]
2023-03-31T00:00:00
[ [ "Li", "Xiaodan", "" ], [ "Chen", "Yuefeng", "" ], [ "Zhu", "Yao", "" ], [ "Wang", "Shuhui", "" ], [ "Zhang", "Rong", "" ], [ "Xue", "Hui", "" ] ]
TITLE: ImageNet-E: Benchmarking Neural Network Robustness via Attribute Editing ABSTRACT: Recent studies have shown that higher accuracy on ImageNet usually leads to better robustness against different corruptions. Therefore, in this paper, instead of following the traditional research paradigm that investigates new out-of-distribution corruptions or perturbations deep models may encounter, we conduct model debugging in in-distribution data to explore which object attributes a model may be sensitive to. To achieve this goal, we create a toolkit for object editing with controls of backgrounds, sizes, positions, and directions, and create a rigorous benchmark named ImageNet-E(diting) for evaluating the image classifier robustness in terms of object attributes. With our ImageNet-E, we evaluate the performance of current deep learning models, including both convolutional neural networks and vision transformers. We find that most models are quite sensitive to attribute changes. A small change in the background can lead to an average of 9.23\% drop on top-1 accuracy. We also evaluate some robust models including both adversarially trained models and other robust trained models and find that some models show worse robustness against attribute changes than vanilla models. Based on these findings, we discover ways to enhance attribute robustness with preprocessing, architecture designs, and training strategies. We hope this work can provide some insights to the community and open up a new avenue for research in robust computer vision. The code and dataset are available at https://github.com/alibaba/easyrobust.
no_new_dataset
0.579698
2303.17111
Xiao Guo
Xiao Guo, Xiaohong Liu, Zhiyuan Ren, Steven Grosz, Iacopo Masi, Xiaoming Liu
Hierarchical Fine-Grained Image Forgery Detection and Localization
To appear at CVPR2023; 17 pages, 15 figures and 10 tables
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Differences in forgery attributes of images generated in CNN-synthesized and image-editing domains are large, and such differences make a unified image forgery detection and localization (IFDL) challenging. To this end, we present a hierarchical fine-grained formulation for IFDL representation learning. Specifically, we first represent forgery attributes of a manipulated image with multiple labels at different levels. Then we perform fine-grained classification at these levels using the hierarchical dependency between them. As a result, the algorithm is encouraged to learn both comprehensive features and inherent hierarchical nature of different forgery attributes, thereby improving the IFDL representation. Our proposed IFDL framework contains three components: multi-branch feature extractor, localization and classification modules. Each branch of the feature extractor learns to classify forgery attributes at one level, while localization and classification modules segment the pixel-level forgery region and detect image-level forgery, respectively. Lastly, we construct a hierarchical fine-grained dataset to facilitate our study. We demonstrate the effectiveness of our method on $7$ different benchmarks, for both tasks of IFDL and forgery attribute classification. Our source code and dataset can be found: \href{https://github.com/CHELSEA234/HiFi_IFDL}{github.com/CHELSEA234/HiFi-IFDL}.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 02:51:52 GMT" } ]
2023-03-31T00:00:00
[ [ "Guo", "Xiao", "" ], [ "Liu", "Xiaohong", "" ], [ "Ren", "Zhiyuan", "" ], [ "Grosz", "Steven", "" ], [ "Masi", "Iacopo", "" ], [ "Liu", "Xiaoming", "" ] ]
TITLE: Hierarchical Fine-Grained Image Forgery Detection and Localization ABSTRACT: Differences in forgery attributes of images generated in CNN-synthesized and image-editing domains are large, and such differences make a unified image forgery detection and localization (IFDL) challenging. To this end, we present a hierarchical fine-grained formulation for IFDL representation learning. Specifically, we first represent forgery attributes of a manipulated image with multiple labels at different levels. Then we perform fine-grained classification at these levels using the hierarchical dependency between them. As a result, the algorithm is encouraged to learn both comprehensive features and inherent hierarchical nature of different forgery attributes, thereby improving the IFDL representation. Our proposed IFDL framework contains three components: multi-branch feature extractor, localization and classification modules. Each branch of the feature extractor learns to classify forgery attributes at one level, while localization and classification modules segment the pixel-level forgery region and detect image-level forgery, respectively. Lastly, we construct a hierarchical fine-grained dataset to facilitate our study. We demonstrate the effectiveness of our method on $7$ different benchmarks, for both tasks of IFDL and forgery attribute classification. Our source code and dataset can be found: \href{https://github.com/CHELSEA234/HiFi_IFDL}{github.com/CHELSEA234/HiFi-IFDL}.
new_dataset
0.715194
2303.17119
Dongsheng Chen
Hao An, Dongsheng Chen, Weiyuan Xu, Zhihong Zhu, Yuexian Zou
TLAG: An Informative Trigger and Label-Aware Knowledge Guided Model for Dialogue-based Relation Extraction
Accepted by CSCWD 2023
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Dialogue-based Relation Extraction (DRE) aims to predict the relation type of argument pairs that are mentioned in dialogue. The latest trigger-enhanced methods propose trigger prediction tasks to promote DRE. However, these methods are not able to fully leverage the trigger information and even bring noise to relation extraction. To solve these problems, we propose TLAG, which fully leverages the trigger and label-aware knowledge to guide the relation extraction. First, we design an adaptive trigger fusion module to fully leverage the trigger information. Then, we introduce label-aware knowledge to further promote our model's performance. Experimental results on the DialogRE dataset show that our TLAG outperforms the baseline models, and detailed analyses demonstrate the effectiveness of our approach.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 03:10:28 GMT" } ]
2023-03-31T00:00:00
[ [ "An", "Hao", "" ], [ "Chen", "Dongsheng", "" ], [ "Xu", "Weiyuan", "" ], [ "Zhu", "Zhihong", "" ], [ "Zou", "Yuexian", "" ] ]
TITLE: TLAG: An Informative Trigger and Label-Aware Knowledge Guided Model for Dialogue-based Relation Extraction ABSTRACT: Dialogue-based Relation Extraction (DRE) aims to predict the relation type of argument pairs that are mentioned in dialogue. The latest trigger-enhanced methods propose trigger prediction tasks to promote DRE. However, these methods are not able to fully leverage the trigger information and even bring noise to relation extraction. To solve these problems, we propose TLAG, which fully leverages the trigger and label-aware knowledge to guide the relation extraction. First, we design an adaptive trigger fusion module to fully leverage the trigger information. Then, we introduce label-aware knowledge to further promote our model's performance. Experimental results on the DialogRE dataset show that our TLAG outperforms the baseline models, and detailed analyses demonstrate the effectiveness of our approach.
no_new_dataset
0.708767
2303.17127
Thalaiyasingam Ajanthan
Thalaiyasingam Ajanthan, Matt Ma, Anton van den Hengel, Stephen Gould
Adaptive Cross Batch Normalization for Metric Learning
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Metric learning is a fundamental problem in computer vision whereby a model is trained to learn a semantically useful embedding space via ranking losses. Traditionally, the effectiveness of a ranking loss depends on the minibatch size, and is, therefore, inherently limited by the memory constraints of the underlying hardware. While simply accumulating the embeddings across minibatches has proved useful (Wang et al. [2020]), we show that it is equally important to ensure that the accumulated embeddings are up to date. In particular, it is necessary to circumvent the representational drift between the accumulated embeddings and the feature embeddings at the current training iteration as the learnable parameters are being updated. In this paper, we model representational drift as distribution misalignment and tackle it using moment matching. The result is a simple method for updating the stored embeddings to match the first and second moments of the current embeddings at each training iteration. Experiments on three popular image retrieval datasets, namely, SOP, In-Shop, and DeepFashion2, demonstrate that our approach significantly improves the performance in all scenarios.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 03:22:52 GMT" } ]
2023-03-31T00:00:00
[ [ "Ajanthan", "Thalaiyasingam", "" ], [ "Ma", "Matt", "" ], [ "Hengel", "Anton van den", "" ], [ "Gould", "Stephen", "" ] ]
TITLE: Adaptive Cross Batch Normalization for Metric Learning ABSTRACT: Metric learning is a fundamental problem in computer vision whereby a model is trained to learn a semantically useful embedding space via ranking losses. Traditionally, the effectiveness of a ranking loss depends on the minibatch size, and is, therefore, inherently limited by the memory constraints of the underlying hardware. While simply accumulating the embeddings across minibatches has proved useful (Wang et al. [2020]), we show that it is equally important to ensure that the accumulated embeddings are up to date. In particular, it is necessary to circumvent the representational drift between the accumulated embeddings and the feature embeddings at the current training iteration as the learnable parameters are being updated. In this paper, we model representational drift as distribution misalignment and tackle it using moment matching. The result is a simple method for updating the stored embeddings to match the first and second moments of the current embeddings at each training iteration. Experiments on three popular image retrieval datasets, namely, SOP, In-Shop, and DeepFashion2, demonstrate that our approach significantly improves the performance in all scenarios.
no_new_dataset
0.709585
2303.17142
Chongjian Ge
Chongjian Ge, Jiangliu Wang, Zhan Tong, Shoufa Chen, Yibing Song, and Ping Luo
Soft Neighbors are Positive Supporters in Contrastive Visual Representation Learning
Accepted by ICLR23
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Contrastive learning methods train visual encoders by comparing views from one instance to others. Typically, the views created from one instance are set as positive, while views from other instances are negative. This binary instance discrimination is studied extensively to improve feature representations in self-supervised learning. In this paper, we rethink the instance discrimination framework and find the binary instance labeling insufficient to measure correlations between different samples. For an intuitive example, given a random image instance, there may exist other images in a mini-batch whose content meanings are the same (i.e., belonging to the same category) or partially related (i.e., belonging to a similar category). How to treat the images that correlate similarly to the current image instance leaves an unexplored problem. We thus propose to support the current image by exploring other correlated instances (i.e., soft neighbors). We first carefully cultivate a candidate neighbor set, which will be further utilized to explore the highly-correlated instances. A cross-attention module is then introduced to predict the correlation score (denoted as positiveness) of other correlated instances with respect to the current one. The positiveness score quantitatively measures the positive support from each correlated instance, and is encoded into the objective for pretext training. To this end, our proposed method benefits in discriminating uncorrelated instances while absorbing correlated instances for SSL. We evaluate our soft neighbor contrastive learning method (SNCLR) on standard visual recognition benchmarks, including image classification, object detection, and instance segmentation. The state-of-the-art recognition performance shows that SNCLR is effective in improving feature representations from both ViT and CNN encoders.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 04:22:07 GMT" } ]
2023-03-31T00:00:00
[ [ "Ge", "Chongjian", "" ], [ "Wang", "Jiangliu", "" ], [ "Tong", "Zhan", "" ], [ "Chen", "Shoufa", "" ], [ "Song", "Yibing", "" ], [ "Luo", "Ping", "" ] ]
TITLE: Soft Neighbors are Positive Supporters in Contrastive Visual Representation Learning ABSTRACT: Contrastive learning methods train visual encoders by comparing views from one instance to others. Typically, the views created from one instance are set as positive, while views from other instances are negative. This binary instance discrimination is studied extensively to improve feature representations in self-supervised learning. In this paper, we rethink the instance discrimination framework and find the binary instance labeling insufficient to measure correlations between different samples. For an intuitive example, given a random image instance, there may exist other images in a mini-batch whose content meanings are the same (i.e., belonging to the same category) or partially related (i.e., belonging to a similar category). How to treat the images that correlate similarly to the current image instance leaves an unexplored problem. We thus propose to support the current image by exploring other correlated instances (i.e., soft neighbors). We first carefully cultivate a candidate neighbor set, which will be further utilized to explore the highly-correlated instances. A cross-attention module is then introduced to predict the correlation score (denoted as positiveness) of other correlated instances with respect to the current one. The positiveness score quantitatively measures the positive support from each correlated instance, and is encoded into the objective for pretext training. To this end, our proposed method benefits in discriminating uncorrelated instances while absorbing correlated instances for SSL. We evaluate our soft neighbor contrastive learning method (SNCLR) on standard visual recognition benchmarks, including image classification, object detection, and instance segmentation. The state-of-the-art recognition performance shows that SNCLR is effective in improving feature representations from both ViT and CNN encoders.
no_new_dataset
0.709799
2303.17147
Yao Yao None
Jingyang Zhang, Yao Yao, Shiwei Li, Jingbo Liu, Tian Fang, David McKinnon, Yanghai Tsin, Long Quan
NeILF++: Inter-Reflectable Light Fields for Geometry and Material Estimation
Project page: \url{https://yoyo000.github.io/NeILF_pp}
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We present a novel differentiable rendering framework for joint geometry, material, and lighting estimation from multi-view images. In contrast to previous methods which assume a simplified environment map or co-located flashlights, in this work, we formulate the lighting of a static scene as one neural incident light field (NeILF) and one outgoing neural radiance field (NeRF). The key insight of the proposed method is the union of the incident and outgoing light fields through physically-based rendering and inter-reflections between surfaces, making it possible to disentangle the scene geometry, material, and lighting from image observations in a physically-based manner. The proposed incident light and inter-reflection framework can be easily applied to other NeRF systems. We show that our method can not only decompose the outgoing radiance into incident lights and surface materials, but also serve as a surface refinement module that further improves the reconstruction detail of the neural surface. We demonstrate on several datasets that the proposed method is able to achieve state-of-the-art results in terms of geometry reconstruction quality, material estimation accuracy, and the fidelity of novel view rendering.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 04:59:48 GMT" } ]
2023-03-31T00:00:00
[ [ "Zhang", "Jingyang", "" ], [ "Yao", "Yao", "" ], [ "Li", "Shiwei", "" ], [ "Liu", "Jingbo", "" ], [ "Fang", "Tian", "" ], [ "McKinnon", "David", "" ], [ "Tsin", "Yanghai", "" ], [ "Quan", "Long", "" ] ]
TITLE: NeILF++: Inter-Reflectable Light Fields for Geometry and Material Estimation ABSTRACT: We present a novel differentiable rendering framework for joint geometry, material, and lighting estimation from multi-view images. In contrast to previous methods which assume a simplified environment map or co-located flashlights, in this work, we formulate the lighting of a static scene as one neural incident light field (NeILF) and one outgoing neural radiance field (NeRF). The key insight of the proposed method is the union of the incident and outgoing light fields through physically-based rendering and inter-reflections between surfaces, making it possible to disentangle the scene geometry, material, and lighting from image observations in a physically-based manner. The proposed incident light and inter-reflection framework can be easily applied to other NeRF systems. We show that our method can not only decompose the outgoing radiance into incident lights and surface materials, but also serve as a surface refinement module that further improves the reconstruction detail of the neural surface. We demonstrate on several datasets that the proposed method is able to achieve state-of-the-art results in terms of geometry reconstruction quality, material estimation accuracy, and the fidelity of novel view rendering.
no_new_dataset
0.712401
2303.17158
Kaiwen Cui
Kaiwen Cui, Yingchen Yu, Fangneng Zhan, Shengcai Liao, Shijian Lu1, Eric Xing
KD-DLGAN: Data Limited Image Generation via Knowledge Distillation
null
CVPR2023
null
null
cs.CV eess.IV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Generative Adversarial Networks (GANs) rely heavily on large-scale training data for training high-quality image generation models. With limited training data, the GAN discriminator often suffers from severe overfitting which directly leads to degraded generation especially in generation diversity. Inspired by the recent advances in knowledge distillation (KD), we propose KD-DLGAN, a knowledge-distillation based generation framework that introduces pre-trained vision-language models for training effective data-limited generation models. KD-DLGAN consists of two innovative designs. The first is aggregated generative KD that mitigates the discriminator overfitting by challenging the discriminator with harder learning tasks and distilling more generalizable knowledge from the pre-trained models. The second is correlated generative KD that improves the generation diversity by distilling and preserving the diverse image-text correlation within the pre-trained models. Extensive experiments over multiple benchmarks show that KD-DLGAN achieves superior image generation with limited training data. In addition, KD-DLGAN complements the state-of-the-art with consistent and substantial performance gains.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 05:36:06 GMT" } ]
2023-03-31T00:00:00
[ [ "Cui", "Kaiwen", "" ], [ "Yu", "Yingchen", "" ], [ "Zhan", "Fangneng", "" ], [ "Liao", "Shengcai", "" ], [ "Lu1", "Shijian", "" ], [ "Xing", "Eric", "" ] ]
TITLE: KD-DLGAN: Data Limited Image Generation via Knowledge Distillation ABSTRACT: Generative Adversarial Networks (GANs) rely heavily on large-scale training data for training high-quality image generation models. With limited training data, the GAN discriminator often suffers from severe overfitting which directly leads to degraded generation especially in generation diversity. Inspired by the recent advances in knowledge distillation (KD), we propose KD-DLGAN, a knowledge-distillation based generation framework that introduces pre-trained vision-language models for training effective data-limited generation models. KD-DLGAN consists of two innovative designs. The first is aggregated generative KD that mitigates the discriminator overfitting by challenging the discriminator with harder learning tasks and distilling more generalizable knowledge from the pre-trained models. The second is correlated generative KD that improves the generation diversity by distilling and preserving the diverse image-text correlation within the pre-trained models. Extensive experiments over multiple benchmarks show that KD-DLGAN achieves superior image generation with limited training data. In addition, KD-DLGAN complements the state-of-the-art with consistent and substantial performance gains.
no_new_dataset
0.711844
2303.17161
Yu Wang
Sid Wang, Akshat Shrivastava, Sasha Livshits
TreePiece: Faster Semantic Parsing via Tree Tokenization
4 pages main body + 4 pages appendices
null
null
null
cs.CL cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Autoregressive (AR) encoder-decoder neural networks have proved successful in many NLP problems, including Semantic Parsing -- a task that translates natural language to machine-readable parse trees. However, the sequential prediction process of AR models can be slow. To accelerate AR for semantic parsing, we introduce a new technique called TreePiece that tokenizes a parse tree into subtrees and generates one subtree per decoding step. On TopV2 benchmark, TreePiece shows 4.6 times faster decoding speed than standard AR, and comparable speed but significantly higher accuracy compared to Non-Autoregressive (NAR).
[ { "version": "v1", "created": "Thu, 30 Mar 2023 05:44:44 GMT" } ]
2023-03-31T00:00:00
[ [ "Wang", "Sid", "" ], [ "Shrivastava", "Akshat", "" ], [ "Livshits", "Sasha", "" ] ]
TITLE: TreePiece: Faster Semantic Parsing via Tree Tokenization ABSTRACT: Autoregressive (AR) encoder-decoder neural networks have proved successful in many NLP problems, including Semantic Parsing -- a task that translates natural language to machine-readable parse trees. However, the sequential prediction process of AR models can be slow. To accelerate AR for semantic parsing, we introduce a new technique called TreePiece that tokenizes a parse tree into subtrees and generates one subtree per decoding step. On TopV2 benchmark, TreePiece shows 4.6 times faster decoding speed than standard AR, and comparable speed but significantly higher accuracy compared to Non-Autoregressive (NAR).
no_new_dataset
0.710069
2303.17167
Jingjing Wang
Hang Du, Xuejun Yan, Jingjing Wang, Di Xie, Shiliang Pu
Rethinking the Approximation Error in 3D Surface Fitting for Point Cloud Normal Estimation
The first two authors contributed equally to this work. The source code are available at https://github.com/hikvision-research/3DVision. Accepted to CVPR 2023
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Most existing approaches for point cloud normal estimation aim to locally fit a geometric surface and calculate the normal from the fitted surface. Recently, learning-based methods have adopted a routine of predicting point-wise weights to solve the weighted least-squares surface fitting problem. Despite achieving remarkable progress, these methods overlook the approximation error of the fitting problem, resulting in a less accurate fitted surface. In this paper, we first carry out in-depth analysis of the approximation error in the surface fitting problem. Then, in order to bridge the gap between estimated and precise surface normals, we present two basic design principles: 1) applies the $Z$-direction Transform to rotate local patches for a better surface fitting with a lower approximation error; 2) models the error of the normal estimation as a learnable term. We implement these two principles using deep neural networks, and integrate them with the state-of-the-art (SOTA) normal estimation methods in a plug-and-play manner. Extensive experiments verify our approaches bring benefits to point cloud normal estimation and push the frontier of state-of-the-art performance on both synthetic and real-world datasets.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 05:59:43 GMT" } ]
2023-03-31T00:00:00
[ [ "Du", "Hang", "" ], [ "Yan", "Xuejun", "" ], [ "Wang", "Jingjing", "" ], [ "Xie", "Di", "" ], [ "Pu", "Shiliang", "" ] ]
TITLE: Rethinking the Approximation Error in 3D Surface Fitting for Point Cloud Normal Estimation ABSTRACT: Most existing approaches for point cloud normal estimation aim to locally fit a geometric surface and calculate the normal from the fitted surface. Recently, learning-based methods have adopted a routine of predicting point-wise weights to solve the weighted least-squares surface fitting problem. Despite achieving remarkable progress, these methods overlook the approximation error of the fitting problem, resulting in a less accurate fitted surface. In this paper, we first carry out in-depth analysis of the approximation error in the surface fitting problem. Then, in order to bridge the gap between estimated and precise surface normals, we present two basic design principles: 1) applies the $Z$-direction Transform to rotate local patches for a better surface fitting with a lower approximation error; 2) models the error of the normal estimation as a learnable term. We implement these two principles using deep neural networks, and integrate them with the state-of-the-art (SOTA) normal estimation methods in a plug-and-play manner. Extensive experiments verify our approaches bring benefits to point cloud normal estimation and push the frontier of state-of-the-art performance on both synthetic and real-world datasets.
no_new_dataset
0.711469
2303.17169
Sifan Long
Sifan Long, Zhen Zhao, Junkun Yuan, Zichang Tan, Jiangjiang Liu, Luping Zhou, Shengsheng Wang, Jingdong Wang
Task-Oriented Multi-Modal Mutual Leaning for Vision-Language Models
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Prompt learning has become one of the most efficient paradigms for adapting large pre-trained vision-language models to downstream tasks. Current state-of-the-art methods, like CoOp and ProDA, tend to adopt soft prompts to learn an appropriate prompt for each specific task. Recent CoCoOp further boosts the base-to-new generalization performance via an image-conditional prompt. However, it directly fuses identical image semantics to prompts of different labels and significantly weakens the discrimination among different classes as shown in our experiments. Motivated by this observation, we first propose a class-aware text prompt (CTP) to enrich generated prompts with label-related image information. Unlike CoCoOp, CTP can effectively involve image semantics and avoid introducing extra ambiguities into different prompts. On the other hand, instead of reserving the complete image representations, we propose text-guided feature tuning (TFT) to make the image branch attend to class-related representation. A contrastive loss is employed to align such augmented text and image representations on downstream tasks. In this way, the image-to-text CTP and text-to-image TFT can be mutually promoted to enhance the adaptation of VLMs for downstream tasks. Extensive experiments demonstrate that our method outperforms the existing methods by a significant margin. Especially, compared to CoCoOp, we achieve an average improvement of 4.03% on new classes and 3.19% on harmonic-mean over eleven classification benchmarks.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 06:02:40 GMT" } ]
2023-03-31T00:00:00
[ [ "Long", "Sifan", "" ], [ "Zhao", "Zhen", "" ], [ "Yuan", "Junkun", "" ], [ "Tan", "Zichang", "" ], [ "Liu", "Jiangjiang", "" ], [ "Zhou", "Luping", "" ], [ "Wang", "Shengsheng", "" ], [ "Wang", "Jingdong", "" ] ]
TITLE: Task-Oriented Multi-Modal Mutual Leaning for Vision-Language Models ABSTRACT: Prompt learning has become one of the most efficient paradigms for adapting large pre-trained vision-language models to downstream tasks. Current state-of-the-art methods, like CoOp and ProDA, tend to adopt soft prompts to learn an appropriate prompt for each specific task. Recent CoCoOp further boosts the base-to-new generalization performance via an image-conditional prompt. However, it directly fuses identical image semantics to prompts of different labels and significantly weakens the discrimination among different classes as shown in our experiments. Motivated by this observation, we first propose a class-aware text prompt (CTP) to enrich generated prompts with label-related image information. Unlike CoCoOp, CTP can effectively involve image semantics and avoid introducing extra ambiguities into different prompts. On the other hand, instead of reserving the complete image representations, we propose text-guided feature tuning (TFT) to make the image branch attend to class-related representation. A contrastive loss is employed to align such augmented text and image representations on downstream tasks. In this way, the image-to-text CTP and text-to-image TFT can be mutually promoted to enhance the adaptation of VLMs for downstream tasks. Extensive experiments demonstrate that our method outperforms the existing methods by a significant margin. Especially, compared to CoCoOp, we achieve an average improvement of 4.03% on new classes and 3.19% on harmonic-mean over eleven classification benchmarks.
no_new_dataset
0.707771
2303.17176
Jerone Andrews
Jerone T. A. Andrews, Przemyslaw Joniak, Alice Xiang
A View From Somewhere: Human-Centric Face Representations
Accepted to ICLR 2023. Code and data may be found at https://github.com/SonyAI/a_view_from_somewhere
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Few datasets contain self-identified sensitive attributes, inferring attributes risks introducing additional biases, and collecting attributes can carry legal risks. Besides, categorical labels can fail to reflect the continuous nature of human phenotypic diversity, making it difficult to compare the similarity between same-labeled faces. To address these issues, we present A View From Somewhere (AVFS) -- a dataset of 638,180 human judgments of face similarity. We demonstrate the utility of AVFS for learning a continuous, low-dimensional embedding space aligned with human perception. Our embedding space, induced under a novel conditional framework, not only enables the accurate prediction of face similarity, but also provides a human-interpretable decomposition of the dimensions used in the human-decision making process, and the importance distinct annotators place on each dimension. We additionally show the practicality of the dimensions for collecting continuous attributes, performing classification, and comparing dataset attribute disparities.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 06:24:54 GMT" } ]
2023-03-31T00:00:00
[ [ "Andrews", "Jerone T. A.", "" ], [ "Joniak", "Przemyslaw", "" ], [ "Xiang", "Alice", "" ] ]
TITLE: A View From Somewhere: Human-Centric Face Representations ABSTRACT: Few datasets contain self-identified sensitive attributes, inferring attributes risks introducing additional biases, and collecting attributes can carry legal risks. Besides, categorical labels can fail to reflect the continuous nature of human phenotypic diversity, making it difficult to compare the similarity between same-labeled faces. To address these issues, we present A View From Somewhere (AVFS) -- a dataset of 638,180 human judgments of face similarity. We demonstrate the utility of AVFS for learning a continuous, low-dimensional embedding space aligned with human perception. Our embedding space, induced under a novel conditional framework, not only enables the accurate prediction of face similarity, but also provides a human-interpretable decomposition of the dimensions used in the human-decision making process, and the importance distinct annotators place on each dimension. We additionally show the practicality of the dimensions for collecting continuous attributes, performing classification, and comparing dataset attribute disparities.
new_dataset
0.715821
2303.17183
Magnus Sahlgren
Joey \"Ohman, Severine Verlinden, Ariel Ekgren, Amaru Cuba Gyllensten, Tim Isbister, Evangelia Gogoulou, Fredrik Carlsson, Magnus Sahlgren
The Nordic Pile: A 1.2TB Nordic Dataset for Language Modeling
null
null
null
null
cs.CL cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Pre-training Large Language Models (LLMs) require massive amounts of text data, and the performance of the LLMs typically correlates with the scale and quality of the datasets. This means that it may be challenging to build LLMs for smaller languages such as Nordic ones, where the availability of text corpora is limited. In order to facilitate the development of the LLMS in the Nordic languages, we curate a high-quality dataset consisting of 1.2TB of text, in all of the major North Germanic languages (Danish, Icelandic, Norwegian, and Swedish), as well as some high-quality English data. This paper details our considerations and processes for collecting, cleaning, and filtering the dataset.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 06:42:22 GMT" } ]
2023-03-31T00:00:00
[ [ "Öhman", "Joey", "" ], [ "Verlinden", "Severine", "" ], [ "Ekgren", "Ariel", "" ], [ "Gyllensten", "Amaru Cuba", "" ], [ "Isbister", "Tim", "" ], [ "Gogoulou", "Evangelia", "" ], [ "Carlsson", "Fredrik", "" ], [ "Sahlgren", "Magnus", "" ] ]
TITLE: The Nordic Pile: A 1.2TB Nordic Dataset for Language Modeling ABSTRACT: Pre-training Large Language Models (LLMs) require massive amounts of text data, and the performance of the LLMs typically correlates with the scale and quality of the datasets. This means that it may be challenging to build LLMs for smaller languages such as Nordic ones, where the availability of text corpora is limited. In order to facilitate the development of the LLMS in the Nordic languages, we curate a high-quality dataset consisting of 1.2TB of text, in all of the major North Germanic languages (Danish, Icelandic, Norwegian, and Swedish), as well as some high-quality English data. This paper details our considerations and processes for collecting, cleaning, and filtering the dataset.
new_dataset
0.715039
2303.17212
Arbish Akram
Arbish Akram and Nazar Khan
SARGAN: Spatial Attention-based Residuals for Facial Expression Manipulation
null
IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2023
10.1109/TCSVT.2023.3255243
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Encoder-decoder based architecture has been widely used in the generator of generative adversarial networks for facial manipulation. However, we observe that the current architecture fails to recover the input image color, rich facial details such as skin color or texture and introduces artifacts as well. In this paper, we present a novel method named SARGAN that addresses the above-mentioned limitations from three perspectives. First, we employed spatial attention-based residual block instead of vanilla residual blocks to properly capture the expression-related features to be changed while keeping the other features unchanged. Second, we exploited a symmetric encoder-decoder network to attend facial features at multiple scales. Third, we proposed to train the complete network with a residual connection which relieves the generator of pressure to generate the input face image thereby producing the desired expression by directly feeding the input image towards the end of the generator. Both qualitative and quantitative experimental results show that our proposed model performs significantly better than state-of-the-art methods. In addition, existing models require much larger datasets for training but their performance degrades on out-of-distribution images. In contrast, SARGAN can be trained on smaller facial expressions datasets, which generalizes well on out-of-distribution images including human photographs, portraits, avatars and statues.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 08:15:18 GMT" } ]
2023-03-31T00:00:00
[ [ "Akram", "Arbish", "" ], [ "Khan", "Nazar", "" ] ]
TITLE: SARGAN: Spatial Attention-based Residuals for Facial Expression Manipulation ABSTRACT: Encoder-decoder based architecture has been widely used in the generator of generative adversarial networks for facial manipulation. However, we observe that the current architecture fails to recover the input image color, rich facial details such as skin color or texture and introduces artifacts as well. In this paper, we present a novel method named SARGAN that addresses the above-mentioned limitations from three perspectives. First, we employed spatial attention-based residual block instead of vanilla residual blocks to properly capture the expression-related features to be changed while keeping the other features unchanged. Second, we exploited a symmetric encoder-decoder network to attend facial features at multiple scales. Third, we proposed to train the complete network with a residual connection which relieves the generator of pressure to generate the input face image thereby producing the desired expression by directly feeding the input image towards the end of the generator. Both qualitative and quantitative experimental results show that our proposed model performs significantly better than state-of-the-art methods. In addition, existing models require much larger datasets for training but their performance degrades on out-of-distribution images. In contrast, SARGAN can be trained on smaller facial expressions datasets, which generalizes well on out-of-distribution images including human photographs, portraits, avatars and statues.
no_new_dataset
0.709799
2303.17216
Xingzhe He
Xingzhe He, Gaurav Bharaj, David Ferman, Helge Rhodin, Pablo Garrido
Few-shot Geometry-Aware Keypoint Localization
CVPR 2023
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition 2023
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Supervised keypoint localization methods rely on large manually labeled image datasets, where objects can deform, articulate, or occlude. However, creating such large keypoint labels is time-consuming and costly, and is often error-prone due to inconsistent labeling. Thus, we desire an approach that can learn keypoint localization with fewer yet consistently annotated images. To this end, we present a novel formulation that learns to localize semantically consistent keypoint definitions, even for occluded regions, for varying object categories. We use a few user-labeled 2D images as input examples, which are extended via self-supervision using a larger unlabeled dataset. Unlike unsupervised methods, the few-shot images act as semantic shape constraints for object localization. Furthermore, we introduce 3D geometry-aware constraints to uplift keypoints, achieving more accurate 2D localization. Our general-purpose formulation paves the way for semantically conditioned generative modeling and attains competitive or state-of-the-art accuracy on several datasets, including human faces, eyes, animals, cars, and never-before-seen mouth interior (teeth) localization tasks, not attempted by the previous few-shot methods. Project page: https://xingzhehe.github.io/FewShot3DKP/}{https://xingzhehe.github.io/FewShot3DKP/
[ { "version": "v1", "created": "Thu, 30 Mar 2023 08:19:42 GMT" } ]
2023-03-31T00:00:00
[ [ "He", "Xingzhe", "" ], [ "Bharaj", "Gaurav", "" ], [ "Ferman", "David", "" ], [ "Rhodin", "Helge", "" ], [ "Garrido", "Pablo", "" ] ]
TITLE: Few-shot Geometry-Aware Keypoint Localization ABSTRACT: Supervised keypoint localization methods rely on large manually labeled image datasets, where objects can deform, articulate, or occlude. However, creating such large keypoint labels is time-consuming and costly, and is often error-prone due to inconsistent labeling. Thus, we desire an approach that can learn keypoint localization with fewer yet consistently annotated images. To this end, we present a novel formulation that learns to localize semantically consistent keypoint definitions, even for occluded regions, for varying object categories. We use a few user-labeled 2D images as input examples, which are extended via self-supervision using a larger unlabeled dataset. Unlike unsupervised methods, the few-shot images act as semantic shape constraints for object localization. Furthermore, we introduce 3D geometry-aware constraints to uplift keypoints, achieving more accurate 2D localization. Our general-purpose formulation paves the way for semantically conditioned generative modeling and attains competitive or state-of-the-art accuracy on several datasets, including human faces, eyes, animals, cars, and never-before-seen mouth interior (teeth) localization tasks, not attempted by the previous few-shot methods. Project page: https://xingzhehe.github.io/FewShot3DKP/}{https://xingzhehe.github.io/FewShot3DKP/
no_new_dataset
0.7116
2303.17243
Celia Wafa Ayad
C\'elia Wafa Ayad, Thomas Bonnier, Benjamin Bosch and Jesse Read
Shapley Chains: Extending Shapley Values to Classifier Chains
DS22
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
In spite of increased attention on explainable machine learning models, explaining multi-output predictions has not yet been extensively addressed. Methods that use Shapley values to attribute feature contributions to the decision making are one of the most popular approaches to explain local individual and global predictions. By considering each output separately in multi-output tasks, these methods fail to provide complete feature explanations. We propose Shapley Chains to overcome this issue by including label interdependencies in the explanation design process. Shapley Chains assign Shapley values as feature importance scores in multi-output classification using classifier chains, by separating the direct and indirect influence of these feature scores. Compared to existing methods, this approach allows to attribute a more complete feature contribution to the predictions of multi-output classification tasks. We provide a mechanism to distribute the hidden contributions of the outputs with respect to a given chaining order of these outputs. Moreover, we show how our approach can reveal indirect feature contributions missed by existing approaches. Shapley Chains help to emphasize the real learning factors in multi-output applications and allows a better understanding of the flow of information through output interdependencies in synthetic and real-world datasets.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 09:19:24 GMT" } ]
2023-03-31T00:00:00
[ [ "Ayad", "Célia Wafa", "" ], [ "Bonnier", "Thomas", "" ], [ "Bosch", "Benjamin", "" ], [ "Read", "Jesse", "" ] ]
TITLE: Shapley Chains: Extending Shapley Values to Classifier Chains ABSTRACT: In spite of increased attention on explainable machine learning models, explaining multi-output predictions has not yet been extensively addressed. Methods that use Shapley values to attribute feature contributions to the decision making are one of the most popular approaches to explain local individual and global predictions. By considering each output separately in multi-output tasks, these methods fail to provide complete feature explanations. We propose Shapley Chains to overcome this issue by including label interdependencies in the explanation design process. Shapley Chains assign Shapley values as feature importance scores in multi-output classification using classifier chains, by separating the direct and indirect influence of these feature scores. Compared to existing methods, this approach allows to attribute a more complete feature contribution to the predictions of multi-output classification tasks. We provide a mechanism to distribute the hidden contributions of the outputs with respect to a given chaining order of these outputs. Moreover, we show how our approach can reveal indirect feature contributions missed by existing approaches. Shapley Chains help to emphasize the real learning factors in multi-output applications and allows a better understanding of the flow of information through output interdependencies in synthetic and real-world datasets.
no_new_dataset
0.709806
2303.17247
Yuhang Lu
Yuhang Lu and Touradj Ebrahimi
Impact of Video Processing Operations in Deepfake Detection
null
null
null
null
cs.CV eess.IV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The detection of digital face manipulation in video has attracted extensive attention due to the increased risk to public trust. To counteract the malicious usage of such techniques, deep learning-based deepfake detection methods have been developed and have shown impressive results. However, the performance of these detectors is often evaluated using benchmarks that hardly reflect real-world situations. For example, the impact of various video processing operations on detection accuracy has not been systematically assessed. To address this gap, this paper first analyzes numerous real-world influencing factors and typical video processing operations. Then, a more systematic assessment methodology is proposed, which allows for a quantitative evaluation of a detector's robustness under the influence of different processing operations. Moreover, substantial experiments have been carried out on three popular deepfake detectors, which give detailed analyses on the impact of each operation and bring insights to foster future research.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 09:24:17 GMT" } ]
2023-03-31T00:00:00
[ [ "Lu", "Yuhang", "" ], [ "Ebrahimi", "Touradj", "" ] ]
TITLE: Impact of Video Processing Operations in Deepfake Detection ABSTRACT: The detection of digital face manipulation in video has attracted extensive attention due to the increased risk to public trust. To counteract the malicious usage of such techniques, deep learning-based deepfake detection methods have been developed and have shown impressive results. However, the performance of these detectors is often evaluated using benchmarks that hardly reflect real-world situations. For example, the impact of various video processing operations on detection accuracy has not been systematically assessed. To address this gap, this paper first analyzes numerous real-world influencing factors and typical video processing operations. Then, a more systematic assessment methodology is proposed, which allows for a quantitative evaluation of a detector's robustness under the influence of different processing operations. Moreover, substantial experiments have been carried out on three popular deepfake detectors, which give detailed analyses on the impact of each operation and bring insights to foster future research.
no_new_dataset
0.713987
2303.17264
Ilan Naiman
Nimrod Berman, Ilan Naiman, Omri Azencot
Multifactor Sequential Disentanglement via Structured Koopman Autoencoders
Accepted to ICLR 2023, Notable-top 25% (Spotlight); The first two authors contributed equally
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Disentangling complex data to its latent factors of variation is a fundamental task in representation learning. Existing work on sequential disentanglement mostly provides two factor representations, i.e., it separates the data to time-varying and time-invariant factors. In contrast, we consider multifactor disentanglement in which multiple (more than two) semantic disentangled components are generated. Key to our approach is a strong inductive bias where we assume that the underlying dynamics can be represented linearly in the latent space. Under this assumption, it becomes natural to exploit the recently introduced Koopman autoencoder models. However, disentangled representations are not guaranteed in Koopman approaches, and thus we propose a novel spectral loss term which leads to structured Koopman matrices and disentanglement. Overall, we propose a simple and easy to code new deep model that is fully unsupervised and it supports multifactor disentanglement. We showcase new disentangling abilities such as swapping of individual static factors between characters, and an incremental swap of disentangled factors from the source to the target. Moreover, we evaluate our method extensively on two factor standard benchmark tasks where we significantly improve over competing unsupervised approaches, and we perform competitively in comparison to weakly- and self-supervised state-of-the-art approaches. The code is available at https://github.com/azencot-group/SKD.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 10:01:58 GMT" } ]
2023-03-31T00:00:00
[ [ "Berman", "Nimrod", "" ], [ "Naiman", "Ilan", "" ], [ "Azencot", "Omri", "" ] ]
TITLE: Multifactor Sequential Disentanglement via Structured Koopman Autoencoders ABSTRACT: Disentangling complex data to its latent factors of variation is a fundamental task in representation learning. Existing work on sequential disentanglement mostly provides two factor representations, i.e., it separates the data to time-varying and time-invariant factors. In contrast, we consider multifactor disentanglement in which multiple (more than two) semantic disentangled components are generated. Key to our approach is a strong inductive bias where we assume that the underlying dynamics can be represented linearly in the latent space. Under this assumption, it becomes natural to exploit the recently introduced Koopman autoencoder models. However, disentangled representations are not guaranteed in Koopman approaches, and thus we propose a novel spectral loss term which leads to structured Koopman matrices and disentanglement. Overall, we propose a simple and easy to code new deep model that is fully unsupervised and it supports multifactor disentanglement. We showcase new disentangling abilities such as swapping of individual static factors between characters, and an incremental swap of disentangled factors from the source to the target. Moreover, we evaluate our method extensively on two factor standard benchmark tasks where we significantly improve over competing unsupervised approaches, and we perform competitively in comparison to weakly- and self-supervised state-of-the-art approaches. The code is available at https://github.com/azencot-group/SKD.
no_new_dataset
0.706975
2303.17276
Philipp Koralus
Philipp Koralus, Vincent Wang-Ma\'scianica
Humans in Humans Out: On GPT Converging Toward Common Sense in both Success and Failure
10 pages
null
null
null
cs.AI cs.CL cs.HC cs.LG
http://creativecommons.org/licenses/by/4.0/
Increase in computational scale and fine-tuning has seen a dramatic improvement in the quality of outputs of large language models (LLMs) like GPT. Given that both GPT-3 and GPT-4 were trained on large quantities of human-generated text, we might ask to what extent their outputs reflect patterns of human thinking, both for correct and incorrect cases. The Erotetic Theory of Reason (ETR) provides a symbolic generative model of both human success and failure in thinking, across propositional, quantified, and probabilistic reasoning, as well as decision-making. We presented GPT-3, GPT-3.5, and GPT-4 with 61 central inference and judgment problems from a recent book-length presentation of ETR, consisting of experimentally verified data-points on human judgment and extrapolated data-points predicted by ETR, with correct inference patterns as well as fallacies and framing effects (the ETR61 benchmark). ETR61 includes classics like Wason's card task, illusory inferences, the decoy effect, and opportunity-cost neglect, among others. GPT-3 showed evidence of ETR-predicted outputs for 59% of these examples, rising to 77% in GPT-3.5 and 75% in GPT-4. Remarkably, the production of human-like fallacious judgments increased from 18% in GPT-3 to 33% in GPT-3.5 and 34% in GPT-4. This suggests that larger and more advanced LLMs may develop a tendency toward more human-like mistakes, as relevant thought patterns are inherent in human-produced training data. According to ETR, the same fundamental patterns are involved both in successful and unsuccessful ordinary reasoning, so that the "bad" cases could paradoxically be learned from the "good" cases. We further present preliminary evidence that ETR-inspired prompt engineering could reduce instances of these mistakes.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 10:32:18 GMT" } ]
2023-03-31T00:00:00
[ [ "Koralus", "Philipp", "" ], [ "Wang-Maścianica", "Vincent", "" ] ]
TITLE: Humans in Humans Out: On GPT Converging Toward Common Sense in both Success and Failure ABSTRACT: Increase in computational scale and fine-tuning has seen a dramatic improvement in the quality of outputs of large language models (LLMs) like GPT. Given that both GPT-3 and GPT-4 were trained on large quantities of human-generated text, we might ask to what extent their outputs reflect patterns of human thinking, both for correct and incorrect cases. The Erotetic Theory of Reason (ETR) provides a symbolic generative model of both human success and failure in thinking, across propositional, quantified, and probabilistic reasoning, as well as decision-making. We presented GPT-3, GPT-3.5, and GPT-4 with 61 central inference and judgment problems from a recent book-length presentation of ETR, consisting of experimentally verified data-points on human judgment and extrapolated data-points predicted by ETR, with correct inference patterns as well as fallacies and framing effects (the ETR61 benchmark). ETR61 includes classics like Wason's card task, illusory inferences, the decoy effect, and opportunity-cost neglect, among others. GPT-3 showed evidence of ETR-predicted outputs for 59% of these examples, rising to 77% in GPT-3.5 and 75% in GPT-4. Remarkably, the production of human-like fallacious judgments increased from 18% in GPT-3 to 33% in GPT-3.5 and 34% in GPT-4. This suggests that larger and more advanced LLMs may develop a tendency toward more human-like mistakes, as relevant thought patterns are inherent in human-produced training data. According to ETR, the same fundamental patterns are involved both in successful and unsuccessful ordinary reasoning, so that the "bad" cases could paradoxically be learned from the "good" cases. We further present preliminary evidence that ETR-inspired prompt engineering could reduce instances of these mistakes.
no_new_dataset
0.707657
2303.17285
Pilhyeon Lee
Pilhyeon Lee, Taeoh Kim, Minho Shim, Dongyoon Wee, Hyeran Byun
Decomposed Cross-modal Distillation for RGB-based Temporal Action Detection
Accepted by CVPR 2023
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Temporal action detection aims to predict the time intervals and the classes of action instances in the video. Despite the promising performance, existing two-stream models exhibit slow inference speed due to their reliance on computationally expensive optical flow. In this paper, we introduce a decomposed cross-modal distillation framework to build a strong RGB-based detector by transferring knowledge of the motion modality. Specifically, instead of direct distillation, we propose to separately learn RGB and motion representations, which are in turn combined to perform action localization. The dual-branch design and the asymmetric training objectives enable effective motion knowledge transfer while preserving RGB information intact. In addition, we introduce a local attentive fusion to better exploit the multimodal complementarity. It is designed to preserve the local discriminability of the features that is important for action localization. Extensive experiments on the benchmarks verify the effectiveness of the proposed method in enhancing RGB-based action detectors. Notably, our framework is agnostic to backbones and detection heads, bringing consistent gains across different model combinations.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 10:47:26 GMT" } ]
2023-03-31T00:00:00
[ [ "Lee", "Pilhyeon", "" ], [ "Kim", "Taeoh", "" ], [ "Shim", "Minho", "" ], [ "Wee", "Dongyoon", "" ], [ "Byun", "Hyeran", "" ] ]
TITLE: Decomposed Cross-modal Distillation for RGB-based Temporal Action Detection ABSTRACT: Temporal action detection aims to predict the time intervals and the classes of action instances in the video. Despite the promising performance, existing two-stream models exhibit slow inference speed due to their reliance on computationally expensive optical flow. In this paper, we introduce a decomposed cross-modal distillation framework to build a strong RGB-based detector by transferring knowledge of the motion modality. Specifically, instead of direct distillation, we propose to separately learn RGB and motion representations, which are in turn combined to perform action localization. The dual-branch design and the asymmetric training objectives enable effective motion knowledge transfer while preserving RGB information intact. In addition, we introduce a local attentive fusion to better exploit the multimodal complementarity. It is designed to preserve the local discriminability of the features that is important for action localization. Extensive experiments on the benchmarks verify the effectiveness of the proposed method in enhancing RGB-based action detectors. Notably, our framework is agnostic to backbones and detection heads, bringing consistent gains across different model combinations.
no_new_dataset
0.709044
2303.17294
Yifu Liu
Yifu Liu, Xiaoxia Li, Zhiling Luo, Wei Zhou
JCDNet: Joint of Common and Definite phases Network for Weakly Supervised Temporal Action Localization
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Weakly-supervised temporal action localization aims to localize action instances in untrimmed videos with only video-level supervision. We witness that different actions record common phases, e.g., the run-up in the HighJump and LongJump. These different actions are defined as conjoint actions, whose rest parts are definite phases, e.g., leaping over the bar in a HighJump. Compared with the common phases, the definite phases are more easily localized in existing researches. Most of them formulate this task as a Multiple Instance Learning paradigm, in which the common phases are tended to be confused with the background, and affect the localization completeness of the conjoint actions. To tackle this challenge, we propose a Joint of Common and Definite phases Network (JCDNet) by improving feature discriminability of the conjoint actions. Specifically, we design a Class-Aware Discriminative module to enhance the contribution of the common phases in classification by the guidance of the coarse definite-phase features. Besides, we introduce a temporal attention module to learn robust action-ness scores via modeling temporal dependencies, distinguishing the common phases from the background. Extensive experiments on three datasets (THUMOS14, ActivityNetv1.2, and a conjoint-action subset) demonstrate that JCDNet achieves competitive performance against the state-of-the-art methods. Keywords: weakly-supervised learning, temporal action localization, conjoint action
[ { "version": "v1", "created": "Thu, 30 Mar 2023 11:09:02 GMT" } ]
2023-03-31T00:00:00
[ [ "Liu", "Yifu", "" ], [ "Li", "Xiaoxia", "" ], [ "Luo", "Zhiling", "" ], [ "Zhou", "Wei", "" ] ]
TITLE: JCDNet: Joint of Common and Definite phases Network for Weakly Supervised Temporal Action Localization ABSTRACT: Weakly-supervised temporal action localization aims to localize action instances in untrimmed videos with only video-level supervision. We witness that different actions record common phases, e.g., the run-up in the HighJump and LongJump. These different actions are defined as conjoint actions, whose rest parts are definite phases, e.g., leaping over the bar in a HighJump. Compared with the common phases, the definite phases are more easily localized in existing researches. Most of them formulate this task as a Multiple Instance Learning paradigm, in which the common phases are tended to be confused with the background, and affect the localization completeness of the conjoint actions. To tackle this challenge, we propose a Joint of Common and Definite phases Network (JCDNet) by improving feature discriminability of the conjoint actions. Specifically, we design a Class-Aware Discriminative module to enhance the contribution of the common phases in classification by the guidance of the coarse definite-phase features. Besides, we introduce a temporal attention module to learn robust action-ness scores via modeling temporal dependencies, distinguishing the common phases from the background. Extensive experiments on three datasets (THUMOS14, ActivityNetv1.2, and a conjoint-action subset) demonstrate that JCDNet achieves competitive performance against the state-of-the-art methods. Keywords: weakly-supervised learning, temporal action localization, conjoint action
no_new_dataset
0.709573
2303.17318
Edward Henderson
Edward G. A. Henderson, Marcel van Herk, Eliana M. Vasquez Osorio
The impact of training dataset size and ensemble inference strategies on head and neck auto-segmentation
Accepted in 20th IEEE International Symposium on Biomedical Imaging (ISBI 2023)
null
null
null
cs.CV cs.AI
http://creativecommons.org/licenses/by/4.0/
Convolutional neural networks (CNNs) are increasingly being used to automate segmentation of organs-at-risk in radiotherapy. Since large sets of highly curated data are scarce, we investigated how much data is required to train accurate and robust head and neck auto-segmentation models. For this, an established 3D CNN was trained from scratch with different sized datasets (25-1000 scans) to segment the brainstem, parotid glands and spinal cord in CTs. Additionally, we evaluated multiple ensemble techniques to improve the performance of these models. The segmentations improved with training set size up to 250 scans and the ensemble methods significantly improved performance for all organs. The impact of the ensemble methods was most notable in the smallest datasets, demonstrating their potential for use in cases where large training datasets are difficult to obtain.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 12:14:07 GMT" } ]
2023-03-31T00:00:00
[ [ "Henderson", "Edward G. A.", "" ], [ "van Herk", "Marcel", "" ], [ "Osorio", "Eliana M. Vasquez", "" ] ]
TITLE: The impact of training dataset size and ensemble inference strategies on head and neck auto-segmentation ABSTRACT: Convolutional neural networks (CNNs) are increasingly being used to automate segmentation of organs-at-risk in radiotherapy. Since large sets of highly curated data are scarce, we investigated how much data is required to train accurate and robust head and neck auto-segmentation models. For this, an established 3D CNN was trained from scratch with different sized datasets (25-1000 scans) to segment the brainstem, parotid glands and spinal cord in CTs. Additionally, we evaluated multiple ensemble techniques to improve the performance of these models. The segmentations improved with training set size up to 250 scans and the ensemble methods significantly improved performance for all organs. The impact of the ensemble methods was most notable in the smallest datasets, demonstrating their potential for use in cases where large training datasets are difficult to obtain.
no_new_dataset
0.711469
2303.17324
Anton Thielmann
Anton Thielmann, Quentin Seifert, Arik Reuter, Elisabeth Bergherr, Benjamin S\"afken
Topics in the Haystack: Extracting and Evaluating Topics beyond Coherence
null
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Extracting and identifying latent topics in large text corpora has gained increasing importance in Natural Language Processing (NLP). Most models, whether probabilistic models similar to Latent Dirichlet Allocation (LDA) or neural topic models, follow the same underlying approach of topic interpretability and topic extraction. We propose a method that incorporates a deeper understanding of both sentence and document themes, and goes beyond simply analyzing word frequencies in the data. This allows our model to detect latent topics that may include uncommon words or neologisms, as well as words not present in the documents themselves. Additionally, we propose several new evaluation metrics based on intruder words and similarity measures in the semantic space. We present correlation coefficients with human identification of intruder words and achieve near-human level results at the word-intrusion task. We demonstrate the competitive performance of our method with a large benchmark study, and achieve superior results compared to state-of-the-art topic modeling and document clustering models.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 12:24:25 GMT" } ]
2023-03-31T00:00:00
[ [ "Thielmann", "Anton", "" ], [ "Seifert", "Quentin", "" ], [ "Reuter", "Arik", "" ], [ "Bergherr", "Elisabeth", "" ], [ "Säfken", "Benjamin", "" ] ]
TITLE: Topics in the Haystack: Extracting and Evaluating Topics beyond Coherence ABSTRACT: Extracting and identifying latent topics in large text corpora has gained increasing importance in Natural Language Processing (NLP). Most models, whether probabilistic models similar to Latent Dirichlet Allocation (LDA) or neural topic models, follow the same underlying approach of topic interpretability and topic extraction. We propose a method that incorporates a deeper understanding of both sentence and document themes, and goes beyond simply analyzing word frequencies in the data. This allows our model to detect latent topics that may include uncommon words or neologisms, as well as words not present in the documents themselves. Additionally, we propose several new evaluation metrics based on intruder words and similarity measures in the semantic space. We present correlation coefficients with human identification of intruder words and achieve near-human level results at the word-intrusion task. We demonstrate the competitive performance of our method with a large benchmark study, and achieve superior results compared to state-of-the-art topic modeling and document clustering models.
no_new_dataset
0.709837
2303.17334
Xinxin Hu
Xinxin Hu, Haotian Chen, Junjie Zhang, Hongchang Chen, Shuxin Liu, Xing Li, Yahui Wang, and Xiangyang Xue
GAT-COBO: Cost-Sensitive Graph Neural Network for Telecom Fraud Detection
null
null
null
null
cs.LG cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Along with the rapid evolution of mobile communication technologies, such as 5G, there has been a drastically increase in telecom fraud, which significantly dissipates individual fortune and social wealth. In recent years, graph mining techniques are gradually becoming a mainstream solution for detecting telecom fraud. However, the graph imbalance problem, caused by the Pareto principle, brings severe challenges to graph data mining. This is a new and challenging problem, but little previous work has been noticed. In this paper, we propose a Graph ATtention network with COst-sensitive BOosting (GAT-COBO) for the graph imbalance problem. First, we design a GAT-based base classifier to learn the embeddings of all nodes in the graph. Then, we feed the embeddings into a well-designed cost-sensitive learner for imbalanced learning. Next, we update the weights according to the misclassification cost to make the model focus more on the minority class. Finally, we sum the node embeddings obtained by multiple cost-sensitive learners to obtain a comprehensive node representation, which is used for the downstream anomaly detection task. Extensive experiments on two real-world telecom fraud detection datasets demonstrate that our proposed method is effective for the graph imbalance problem, outperforming the state-of-the-art GNNs and GNN-based fraud detectors. In addition, our model is also helpful for solving the widespread over-smoothing problem in GNNs. The GAT-COBO code and datasets are available at https://github.com/xxhu94/GAT-COBO.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 07:02:50 GMT" } ]
2023-03-31T00:00:00
[ [ "Hu", "Xinxin", "" ], [ "Chen", "Haotian", "" ], [ "Zhang", "Junjie", "" ], [ "Chen", "Hongchang", "" ], [ "Liu", "Shuxin", "" ], [ "Li", "Xing", "" ], [ "Wang", "Yahui", "" ], [ "Xue", "Xiangyang", "" ] ]
TITLE: GAT-COBO: Cost-Sensitive Graph Neural Network for Telecom Fraud Detection ABSTRACT: Along with the rapid evolution of mobile communication technologies, such as 5G, there has been a drastically increase in telecom fraud, which significantly dissipates individual fortune and social wealth. In recent years, graph mining techniques are gradually becoming a mainstream solution for detecting telecom fraud. However, the graph imbalance problem, caused by the Pareto principle, brings severe challenges to graph data mining. This is a new and challenging problem, but little previous work has been noticed. In this paper, we propose a Graph ATtention network with COst-sensitive BOosting (GAT-COBO) for the graph imbalance problem. First, we design a GAT-based base classifier to learn the embeddings of all nodes in the graph. Then, we feed the embeddings into a well-designed cost-sensitive learner for imbalanced learning. Next, we update the weights according to the misclassification cost to make the model focus more on the minority class. Finally, we sum the node embeddings obtained by multiple cost-sensitive learners to obtain a comprehensive node representation, which is used for the downstream anomaly detection task. Extensive experiments on two real-world telecom fraud detection datasets demonstrate that our proposed method is effective for the graph imbalance problem, outperforming the state-of-the-art GNNs and GNN-based fraud detectors. In addition, our model is also helpful for solving the widespread over-smoothing problem in GNNs. The GAT-COBO code and datasets are available at https://github.com/xxhu94/GAT-COBO.
no_new_dataset
0.711638
2303.17358
Chao Xu
Yuxuan Zhang, Chao Xu, Howard H. Yang, Xijun Wang, and Tony Q. S. Quek
DPP-based Client Selection for Federated Learning with Non-IID Data
null
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper proposes a client selection (CS) method to tackle the communication bottleneck of federated learning (FL) while concurrently coping with FL's data heterogeneity issue. Specifically, we first analyze the effect of CS in FL and show that FL training can be accelerated by adequately choosing participants to diversify the training dataset in each round of training. Based on this, we leverage data profiling and determinantal point process (DPP) sampling techniques to develop an algorithm termed Federated Learning with DPP-based Participant Selection (FL-DP$^3$S). This algorithm effectively diversifies the participants' datasets in each round of training while preserving their data privacy. We conduct extensive experiments to examine the efficacy of our proposed method. The results show that our scheme attains a faster convergence rate, as well as a smaller communication overhead than several baselines.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 13:14:54 GMT" } ]
2023-03-31T00:00:00
[ [ "Zhang", "Yuxuan", "" ], [ "Xu", "Chao", "" ], [ "Yang", "Howard H.", "" ], [ "Wang", "Xijun", "" ], [ "Quek", "Tony Q. S.", "" ] ]
TITLE: DPP-based Client Selection for Federated Learning with Non-IID Data ABSTRACT: This paper proposes a client selection (CS) method to tackle the communication bottleneck of federated learning (FL) while concurrently coping with FL's data heterogeneity issue. Specifically, we first analyze the effect of CS in FL and show that FL training can be accelerated by adequately choosing participants to diversify the training dataset in each round of training. Based on this, we leverage data profiling and determinantal point process (DPP) sampling techniques to develop an algorithm termed Federated Learning with DPP-based Participant Selection (FL-DP$^3$S). This algorithm effectively diversifies the participants' datasets in each round of training while preserving their data privacy. We conduct extensive experiments to examine the efficacy of our proposed method. The results show that our scheme attains a faster convergence rate, as well as a smaller communication overhead than several baselines.
no_new_dataset
0.711625
2303.17387
Sudip Mittal
Jesse Ables, Thomas Kirby, Sudip Mittal, Ioana Banicescu, Shahram Rahimi, William Anderson, Maria Seale
Explainable Intrusion Detection Systems Using Competitive Learning Techniques
null
null
null
null
cs.CR cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
The current state of the art systems in Artificial Intelligence (AI) enabled intrusion detection use a variety of black box methods. These black box methods are generally trained using Error Based Learning (EBL) techniques with a focus on creating accurate models. These models have high performative costs and are not easily explainable. A white box Competitive Learning (CL) based eXplainable Intrusion Detection System (X-IDS) offers a potential solution to these problem. CL models utilize an entirely different learning paradigm than EBL approaches. This different learning process makes the CL family of algorithms innately explainable and less resource intensive. In this paper, we create an X-IDS architecture that is based on DARPA's recommendation for explainable systems. In our architecture we leverage CL algorithms like, Self Organizing Maps (SOM), Growing Self Organizing Maps (GSOM), and Growing Hierarchical Self Organizing Map (GHSOM). The resulting models can be data-mined to create statistical and visual explanations. Our architecture is tested using NSL-KDD and CIC-IDS-2017 benchmark datasets, and produces accuracies that are 1% - 3% less than EBL models. However, CL models are much more explainable than EBL models. Additionally, we use a pruning process that is able to significantly reduce the size of these CL based models. By pruning our models, we are able to increase prediction speeds. Lastly, we analyze the statistical and visual explanations generated by our architecture, and we give a strategy that users could use to help navigate the set of explanations. These explanations will help users build trust with an Intrusion Detection System (IDS), and allow users to discover ways to increase the IDS's potency.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 13:58:47 GMT" } ]
2023-03-31T00:00:00
[ [ "Ables", "Jesse", "" ], [ "Kirby", "Thomas", "" ], [ "Mittal", "Sudip", "" ], [ "Banicescu", "Ioana", "" ], [ "Rahimi", "Shahram", "" ], [ "Anderson", "William", "" ], [ "Seale", "Maria", "" ] ]
TITLE: Explainable Intrusion Detection Systems Using Competitive Learning Techniques ABSTRACT: The current state of the art systems in Artificial Intelligence (AI) enabled intrusion detection use a variety of black box methods. These black box methods are generally trained using Error Based Learning (EBL) techniques with a focus on creating accurate models. These models have high performative costs and are not easily explainable. A white box Competitive Learning (CL) based eXplainable Intrusion Detection System (X-IDS) offers a potential solution to these problem. CL models utilize an entirely different learning paradigm than EBL approaches. This different learning process makes the CL family of algorithms innately explainable and less resource intensive. In this paper, we create an X-IDS architecture that is based on DARPA's recommendation for explainable systems. In our architecture we leverage CL algorithms like, Self Organizing Maps (SOM), Growing Self Organizing Maps (GSOM), and Growing Hierarchical Self Organizing Map (GHSOM). The resulting models can be data-mined to create statistical and visual explanations. Our architecture is tested using NSL-KDD and CIC-IDS-2017 benchmark datasets, and produces accuracies that are 1% - 3% less than EBL models. However, CL models are much more explainable than EBL models. Additionally, we use a pruning process that is able to significantly reduce the size of these CL based models. By pruning our models, we are able to increase prediction speeds. Lastly, we analyze the statistical and visual explanations generated by our architecture, and we give a strategy that users could use to help navigate the set of explanations. These explanations will help users build trust with an Intrusion Detection System (IDS), and allow users to discover ways to increase the IDS's potency.
no_new_dataset
0.710635
2303.17388
Linyue Liu
Linyue Liu, Xi Guo, Chun Ouyang, Patrick C. K. Hung, Hong-Yu Zhang, Keqing He, Chen Mo and Zaiwen Feng
BPCE: A Prototype for Co-Evolution between Business Process Variants through Configurable Process Model
18 pages , 11 figures
null
null
null
cs.SE
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
With the continuous development of business process management technology, the increasing business process models are usually owned by large enterprises. In large enterprises, different stakeholders may modify the same business process model. In order to better manage the changeability of processes, they adopt configurable business process models to manage process variants. However, the process variants will vary with the change in enterprise business demands. Therefore, it is necessary to explore the co-evolution of the process variants so as to effectively manage the business process family. To this end, a novel framework for co-evolution between business process variants through a configurable process model is proposed in this work. First, the mapping relationship between process variants and configurable models is standardized in this study. A series of change operations and change propagation operations between process variants and configurable models are further defined for achieving propagation. Then, an overall algorithm is proposed for achieving co-evolution of process variants. Next, a prototype is developed for managing change synchronization between process variants and configurable process models. Finally, the effectiveness and efficiency of our proposed process change propagation method are verified based on experiments on two business process datasets. The experimental results show that our approach implements the co-evolution of process variants with high accuracy and efficiency.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 13:59:34 GMT" } ]
2023-03-31T00:00:00
[ [ "Liu", "Linyue", "" ], [ "Guo", "Xi", "" ], [ "Ouyang", "Chun", "" ], [ "Hung", "Patrick C. K.", "" ], [ "Zhang", "Hong-Yu", "" ], [ "He", "Keqing", "" ], [ "Mo", "Chen", "" ], [ "Feng", "Zaiwen", "" ] ]
TITLE: BPCE: A Prototype for Co-Evolution between Business Process Variants through Configurable Process Model ABSTRACT: With the continuous development of business process management technology, the increasing business process models are usually owned by large enterprises. In large enterprises, different stakeholders may modify the same business process model. In order to better manage the changeability of processes, they adopt configurable business process models to manage process variants. However, the process variants will vary with the change in enterprise business demands. Therefore, it is necessary to explore the co-evolution of the process variants so as to effectively manage the business process family. To this end, a novel framework for co-evolution between business process variants through a configurable process model is proposed in this work. First, the mapping relationship between process variants and configurable models is standardized in this study. A series of change operations and change propagation operations between process variants and configurable models are further defined for achieving propagation. Then, an overall algorithm is proposed for achieving co-evolution of process variants. Next, a prototype is developed for managing change synchronization between process variants and configurable process models. Finally, the effectiveness and efficiency of our proposed process change propagation method are verified based on experiments on two business process datasets. The experimental results show that our approach implements the co-evolution of process variants with high accuracy and efficiency.
no_new_dataset
0.710025
2303.17393
Nan Pu
Nan Pu, Zhun Zhong and Nicu Sebe
Dynamic Conceptional Contrastive Learning for Generalized Category Discovery
10 pages, 5 figures, accepted by CVPR2023
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Generalized category discovery (GCD) is a recently proposed open-world problem, which aims to automatically cluster partially labeled data. The main challenge is that the unlabeled data contain instances that are not only from known categories of the labeled data but also from novel categories. This leads traditional novel category discovery (NCD) methods to be incapacitated for GCD, due to their assumption of unlabeled data are only from novel categories. One effective way for GCD is applying self-supervised learning to learn discriminate representation for unlabeled data. However, this manner largely ignores underlying relationships between instances of the same concepts (e.g., class, super-class, and sub-class), which results in inferior representation learning. In this paper, we propose a Dynamic Conceptional Contrastive Learning (DCCL) framework, which can effectively improve clustering accuracy by alternately estimating underlying visual conceptions and learning conceptional representation. In addition, we design a dynamic conception generation and update mechanism, which is able to ensure consistent conception learning and thus further facilitate the optimization of DCCL. Extensive experiments show that DCCL achieves new state-of-the-art performances on six generic and fine-grained visual recognition datasets, especially on fine-grained ones. For example, our method significantly surpasses the best competitor by 16.2% on the new classes for the CUB-200 dataset. Code is available at https://github.com/TPCD/DCCL.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 14:04:39 GMT" } ]
2023-03-31T00:00:00
[ [ "Pu", "Nan", "" ], [ "Zhong", "Zhun", "" ], [ "Sebe", "Nicu", "" ] ]
TITLE: Dynamic Conceptional Contrastive Learning for Generalized Category Discovery ABSTRACT: Generalized category discovery (GCD) is a recently proposed open-world problem, which aims to automatically cluster partially labeled data. The main challenge is that the unlabeled data contain instances that are not only from known categories of the labeled data but also from novel categories. This leads traditional novel category discovery (NCD) methods to be incapacitated for GCD, due to their assumption of unlabeled data are only from novel categories. One effective way for GCD is applying self-supervised learning to learn discriminate representation for unlabeled data. However, this manner largely ignores underlying relationships between instances of the same concepts (e.g., class, super-class, and sub-class), which results in inferior representation learning. In this paper, we propose a Dynamic Conceptional Contrastive Learning (DCCL) framework, which can effectively improve clustering accuracy by alternately estimating underlying visual conceptions and learning conceptional representation. In addition, we design a dynamic conception generation and update mechanism, which is able to ensure consistent conception learning and thus further facilitate the optimization of DCCL. Extensive experiments show that DCCL achieves new state-of-the-art performances on six generic and fine-grained visual recognition datasets, especially on fine-grained ones. For example, our method significantly surpasses the best competitor by 16.2% on the new classes for the CUB-200 dataset. Code is available at https://github.com/TPCD/DCCL.
no_new_dataset
0.709057
2303.17396
Yicheng Luo
Yicheng Luo, Jackie Kay, Edward Grefenstette, Marc Peter Deisenroth
Finetuning from Offline Reinforcement Learning: Challenges, Trade-offs and Practical Solutions
An abstract of this paper was accepted at RLDM 2022
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Offline reinforcement learning (RL) allows for the training of competent agents from offline datasets without any interaction with the environment. Online finetuning of such offline models can further improve performance. But how should we ideally finetune agents obtained from offline RL training? While offline RL algorithms can in principle be used for finetuning, in practice, their online performance improves slowly. In contrast, we show that it is possible to use standard online off-policy algorithms for faster improvement. However, we find this approach may suffer from policy collapse, where the policy undergoes severe performance deterioration during initial online learning. We investigate the issue of policy collapse and how it relates to data diversity, algorithm choices and online replay distribution. Based on these insights, we propose a conservative policy optimization procedure that can achieve stable and sample-efficient online learning from offline pretraining.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 14:08:31 GMT" } ]
2023-03-31T00:00:00
[ [ "Luo", "Yicheng", "" ], [ "Kay", "Jackie", "" ], [ "Grefenstette", "Edward", "" ], [ "Deisenroth", "Marc Peter", "" ] ]
TITLE: Finetuning from Offline Reinforcement Learning: Challenges, Trade-offs and Practical Solutions ABSTRACT: Offline reinforcement learning (RL) allows for the training of competent agents from offline datasets without any interaction with the environment. Online finetuning of such offline models can further improve performance. But how should we ideally finetune agents obtained from offline RL training? While offline RL algorithms can in principle be used for finetuning, in practice, their online performance improves slowly. In contrast, we show that it is possible to use standard online off-policy algorithms for faster improvement. However, we find this approach may suffer from policy collapse, where the policy undergoes severe performance deterioration during initial online learning. We investigate the issue of policy collapse and how it relates to data diversity, algorithm choices and online replay distribution. Based on these insights, we propose a conservative policy optimization procedure that can achieve stable and sample-efficient online learning from offline pretraining.
no_new_dataset
0.709667
2303.17431
Nejat ARINIK
Nejat Arinik (UMR TETIS, INRAE, LIA), Roberto Interdonato (UMR TETIS, Cirad), Mathieu Roche (UMR TETIS, Cirad, Cirad-ES), Maguelonne Teisseire (UMR TETIS, INRAE)
An evaluation framework for comparing epidemic intelligence systems
null
IEEE Access, 2023, pp.1 - 1
10.1109/ACCESS.2023.3262462
null
cs.IR cs.SI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In the context of Epidemic Intelligence, many Event-Based Surveillance (EBS) systems have been proposed in the literature to promote the early identification and characterization of potential health threats from online sources of any nature. Each EBS system has its own surveillance definitions and priorities, therefore this makes the task of selecting the most appropriate EBS system for a given situation a challenge for end-users. In this work, we propose a new evaluation framework to address this issue. It first transforms the raw input epidemiological event data into a set of normalized events with multi-granularity, then conducts a descriptive retrospective analysis based on four evaluation objectives: spatial, temporal, thematic and source analysis. We illustrate its relevance by applying it to an Avian Influenza dataset collected by a selection of EBS systems, and show how our framework allows identifying their strengths and drawbacks in terms of epidemic surveillance.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 14:55:43 GMT" } ]
2023-03-31T00:00:00
[ [ "Arinik", "Nejat", "", "UMR TETIS, INRAE, LIA" ], [ "Interdonato", "Roberto", "", "UMR TETIS,\n Cirad" ], [ "Roche", "Mathieu", "", "UMR TETIS, Cirad, Cirad-ES" ], [ "Teisseire", "Maguelonne", "", "UMR\n TETIS, INRAE" ] ]
TITLE: An evaluation framework for comparing epidemic intelligence systems ABSTRACT: In the context of Epidemic Intelligence, many Event-Based Surveillance (EBS) systems have been proposed in the literature to promote the early identification and characterization of potential health threats from online sources of any nature. Each EBS system has its own surveillance definitions and priorities, therefore this makes the task of selecting the most appropriate EBS system for a given situation a challenge for end-users. In this work, we propose a new evaluation framework to address this issue. It first transforms the raw input epidemiological event data into a set of normalized events with multi-granularity, then conducts a descriptive retrospective analysis based on four evaluation objectives: spatial, temporal, thematic and source analysis. We illustrate its relevance by applying it to an Avian Influenza dataset collected by a selection of EBS systems, and show how our framework allows identifying their strengths and drawbacks in terms of epidemic surveillance.
no_new_dataset
0.712188
2303.17468
Evelyn Ruff
Evelyn Ruff, Rebecca Russell, Matthew Stoeckle, Piero Miotto, and Jonathan P. How
Surrogate Neural Networks for Efficient Simulation-based Trajectory Planning Optimization
8 pages, 11 figures, submitted to the IEEE Conference of Decision and Control 2023
null
null
null
math.OC cs.LG
http://creativecommons.org/licenses/by/4.0/
This paper presents a novel methodology that uses surrogate models in the form of neural networks to reduce the computation time of simulation-based optimization of a reference trajectory. Simulation-based optimization is necessary when there is no analytical form of the system accessible, only input-output data that can be used to create a surrogate model of the simulation. Like many high-fidelity simulations, this trajectory planning simulation is very nonlinear and computationally expensive, making it challenging to optimize iteratively. Through gradient descent optimization, our approach finds the optimal reference trajectory for landing a hypersonic vehicle. In contrast to the large datasets used to create the surrogate models in prior literature, our methodology is specifically designed to minimize the number of simulation executions required by the gradient descent optimizer. We demonstrated this methodology to be more efficient than the standard practice of hand-tuning the inputs through trial-and-error or randomly sampling the input parameter space. Due to the intelligently selected input values to the simulation, our approach yields better simulation outcomes that are achieved more rapidly and to a higher degree of accuracy. Optimizing the hypersonic vehicle's reference trajectory is very challenging due to the simulation's extreme nonlinearity, but even so, this novel approach found a 74% better-performing reference trajectory compared to nominal, and the numerical results clearly show a substantial reduction in computation time for designing future trajectories.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 15:44:30 GMT" } ]
2023-03-31T00:00:00
[ [ "Ruff", "Evelyn", "" ], [ "Russell", "Rebecca", "" ], [ "Stoeckle", "Matthew", "" ], [ "Miotto", "Piero", "" ], [ "How", "Jonathan P.", "" ] ]
TITLE: Surrogate Neural Networks for Efficient Simulation-based Trajectory Planning Optimization ABSTRACT: This paper presents a novel methodology that uses surrogate models in the form of neural networks to reduce the computation time of simulation-based optimization of a reference trajectory. Simulation-based optimization is necessary when there is no analytical form of the system accessible, only input-output data that can be used to create a surrogate model of the simulation. Like many high-fidelity simulations, this trajectory planning simulation is very nonlinear and computationally expensive, making it challenging to optimize iteratively. Through gradient descent optimization, our approach finds the optimal reference trajectory for landing a hypersonic vehicle. In contrast to the large datasets used to create the surrogate models in prior literature, our methodology is specifically designed to minimize the number of simulation executions required by the gradient descent optimizer. We demonstrated this methodology to be more efficient than the standard practice of hand-tuning the inputs through trial-and-error or randomly sampling the input parameter space. Due to the intelligently selected input values to the simulation, our approach yields better simulation outcomes that are achieved more rapidly and to a higher degree of accuracy. Optimizing the hypersonic vehicle's reference trajectory is very challenging due to the simulation's extreme nonlinearity, but even so, this novel approach found a 74% better-performing reference trajectory compared to nominal, and the numerical results clearly show a substantial reduction in computation time for designing future trajectories.
no_new_dataset
0.711205
2303.17472
Qitao Zhao
Qitao Zhao, Ce Zheng, Mengyuan Liu, Pichao Wang, Chen Chen
PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimation
Accepted to CVPR 2023. Project page: https://qitaozhao.github.io/PoseFormerV2
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recently, transformer-based methods have gained significant success in sequential 2D-to-3D lifting human pose estimation. As a pioneering work, PoseFormer captures spatial relations of human joints in each video frame and human dynamics across frames with cascaded transformer layers and has achieved impressive performance. However, in real scenarios, the performance of PoseFormer and its follow-ups is limited by two factors: (a) The length of the input joint sequence; (b) The quality of 2D joint detection. Existing methods typically apply self-attention to all frames of the input sequence, causing a huge computational burden when the frame number is increased to obtain advanced estimation accuracy, and they are not robust to noise naturally brought by the limited capability of 2D joint detectors. In this paper, we propose PoseFormerV2, which exploits a compact representation of lengthy skeleton sequences in the frequency domain to efficiently scale up the receptive field and boost robustness to noisy 2D joint detection. With minimum modifications to PoseFormer, the proposed method effectively fuses features both in the time domain and frequency domain, enjoying a better speed-accuracy trade-off than its precursor. Extensive experiments on two benchmark datasets (i.e., Human3.6M and MPI-INF-3DHP) demonstrate that the proposed approach significantly outperforms the original PoseFormer and other transformer-based variants. Code is released at \url{https://github.com/QitaoZhao/PoseFormerV2}.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 15:45:51 GMT" } ]
2023-03-31T00:00:00
[ [ "Zhao", "Qitao", "" ], [ "Zheng", "Ce", "" ], [ "Liu", "Mengyuan", "" ], [ "Wang", "Pichao", "" ], [ "Chen", "Chen", "" ] ]
TITLE: PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimation ABSTRACT: Recently, transformer-based methods have gained significant success in sequential 2D-to-3D lifting human pose estimation. As a pioneering work, PoseFormer captures spatial relations of human joints in each video frame and human dynamics across frames with cascaded transformer layers and has achieved impressive performance. However, in real scenarios, the performance of PoseFormer and its follow-ups is limited by two factors: (a) The length of the input joint sequence; (b) The quality of 2D joint detection. Existing methods typically apply self-attention to all frames of the input sequence, causing a huge computational burden when the frame number is increased to obtain advanced estimation accuracy, and they are not robust to noise naturally brought by the limited capability of 2D joint detectors. In this paper, we propose PoseFormerV2, which exploits a compact representation of lengthy skeleton sequences in the frequency domain to efficiently scale up the receptive field and boost robustness to noisy 2D joint detection. With minimum modifications to PoseFormer, the proposed method effectively fuses features both in the time domain and frequency domain, enjoying a better speed-accuracy trade-off than its precursor. Extensive experiments on two benchmark datasets (i.e., Human3.6M and MPI-INF-3DHP) demonstrate that the proposed approach significantly outperforms the original PoseFormer and other transformer-based variants. Code is released at \url{https://github.com/QitaoZhao/PoseFormerV2}.
no_new_dataset
0.709403
2303.17480
Jiadong Wang Mr.
Jiadong Wang, Xinyuan Qian, Malu Zhang, Robby T. Tan, Haizhou Li
Seeing What You Said: Talking Face Generation Guided by a Lip Reading Expert
accepted by CVPR 2023
null
null
null
cs.CV cs.AI eess.IV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Talking face generation, also known as speech-to-lip generation, reconstructs facial motions concerning lips given coherent speech input. The previous studies revealed the importance of lip-speech synchronization and visual quality. Despite much progress, they hardly focus on the content of lip movements i.e., the visual intelligibility of the spoken words, which is an important aspect of generation quality. To address the problem, we propose using a lip-reading expert to improve the intelligibility of the generated lip regions by penalizing the incorrect generation results. Moreover, to compensate for data scarcity, we train the lip-reading expert in an audio-visual self-supervised manner. With a lip-reading expert, we propose a novel contrastive learning to enhance lip-speech synchronization, and a transformer to encode audio synchronically with video, while considering global temporal dependency of audio. For evaluation, we propose a new strategy with two different lip-reading experts to measure intelligibility of the generated videos. Rigorous experiments show that our proposal is superior to other State-of-the-art (SOTA) methods, such as Wav2Lip, in reading intelligibility i.e., over 38% Word Error Rate (WER) on LRS2 dataset and 27.8% accuracy on LRW dataset. We also achieve the SOTA performance in lip-speech synchronization and comparable performances in visual quality.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 07:51:07 GMT" } ]
2023-03-31T00:00:00
[ [ "Wang", "Jiadong", "" ], [ "Qian", "Xinyuan", "" ], [ "Zhang", "Malu", "" ], [ "Tan", "Robby T.", "" ], [ "Li", "Haizhou", "" ] ]
TITLE: Seeing What You Said: Talking Face Generation Guided by a Lip Reading Expert ABSTRACT: Talking face generation, also known as speech-to-lip generation, reconstructs facial motions concerning lips given coherent speech input. The previous studies revealed the importance of lip-speech synchronization and visual quality. Despite much progress, they hardly focus on the content of lip movements i.e., the visual intelligibility of the spoken words, which is an important aspect of generation quality. To address the problem, we propose using a lip-reading expert to improve the intelligibility of the generated lip regions by penalizing the incorrect generation results. Moreover, to compensate for data scarcity, we train the lip-reading expert in an audio-visual self-supervised manner. With a lip-reading expert, we propose a novel contrastive learning to enhance lip-speech synchronization, and a transformer to encode audio synchronically with video, while considering global temporal dependency of audio. For evaluation, we propose a new strategy with two different lip-reading experts to measure intelligibility of the generated videos. Rigorous experiments show that our proposal is superior to other State-of-the-art (SOTA) methods, such as Wav2Lip, in reading intelligibility i.e., over 38% Word Error Rate (WER) on LRS2 dataset and 27.8% accuracy on LRW dataset. We also achieve the SOTA performance in lip-speech synchronization and comparable performances in visual quality.
no_new_dataset
0.711418
2303.17482
Tao Zhang
Xue Zaifa, Lu Huibin, Zhang Tao, Li Tao and Lu Xin
Three-way causal attribute partial order structure analysis
null
null
null
null
cs.AI cs.LG stat.ME
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
As an emerging concept cognitive learning model, partial order formal structure analysis (POFSA) has been widely used in the field of knowledge processing. In this paper, we propose the method named three-way causal attribute partial order structure (3WCAPOS) to evolve the POFSA from set coverage to causal coverage in order to increase the interpretability and classification performance of the model. First, the concept of causal factor (CF) is proposed to evaluate the causal correlation between attributes and decision attributes in the formal decision context. Then, combining CF with attribute partial order structure, the concept of causal attribute partial order structure is defined and makes set coverage evolve into causal coverage. Finally, combined with the idea of three-way decision, 3WCAPOS is formed, which makes the purity of nodes in the structure clearer and the changes between levels more obviously. In addition, the experiments are carried out from the classification ability and the interpretability of the structure through the six datasets. Through these experiments, it is concluded the accuracy of 3WCAPOS is improved by 1% - 9% compared with classification and regression tree, and more interpretable and the processing of knowledge is more reasonable compared with attribute partial order structure.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 03:52:20 GMT" } ]
2023-03-31T00:00:00
[ [ "Zaifa", "Xue", "" ], [ "Huibin", "Lu", "" ], [ "Tao", "Zhang", "" ], [ "Tao", "Li", "" ], [ "Xin", "Lu", "" ] ]
TITLE: Three-way causal attribute partial order structure analysis ABSTRACT: As an emerging concept cognitive learning model, partial order formal structure analysis (POFSA) has been widely used in the field of knowledge processing. In this paper, we propose the method named three-way causal attribute partial order structure (3WCAPOS) to evolve the POFSA from set coverage to causal coverage in order to increase the interpretability and classification performance of the model. First, the concept of causal factor (CF) is proposed to evaluate the causal correlation between attributes and decision attributes in the formal decision context. Then, combining CF with attribute partial order structure, the concept of causal attribute partial order structure is defined and makes set coverage evolve into causal coverage. Finally, combined with the idea of three-way decision, 3WCAPOS is formed, which makes the purity of nodes in the structure clearer and the changes between levels more obviously. In addition, the experiments are carried out from the classification ability and the interpretability of the structure through the six datasets. Through these experiments, it is concluded the accuracy of 3WCAPOS is improved by 1% - 9% compared with classification and regression tree, and more interpretable and the processing of knowledge is more reasonable compared with attribute partial order structure.
no_new_dataset
0.713032
2303.17486
Xinxin Hu
Xinxin Hu, Haotian Chen, Hongchang Chen, Shuxin Liu, Xing Li, Shibo Zhang, Yahui Wang, and Xiangyang Xue
Cost Sensitive GNN-based Imbalanced Learning for Mobile Social Network Fraud Detection
null
null
null
null
cs.SI cs.AI cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
With the rapid development of mobile networks, the people's social contacts have been considerably facilitated. However, the rise of mobile social network fraud upon those networks, has caused a great deal of distress, in case of depleting personal and social wealth, then potentially doing significant economic harm. To detect fraudulent users, call detail record (CDR) data, which portrays the social behavior of users in mobile networks, has been widely utilized. But the imbalance problem in the aforementioned data, which could severely hinder the effectiveness of fraud detectors based on graph neural networks(GNN), has hardly been addressed in previous work. In this paper, we are going to present a novel Cost-Sensitive Graph Neural Network (CSGNN) by creatively combining cost-sensitive learning and graph neural networks. We conduct extensive experiments on two open-source realworld mobile network fraud datasets. The results show that CSGNN can effectively solve the graph imbalance problem and then achieve better detection performance than the state-of-the-art algorithms. We believe that our research can be applied to solve the graph imbalance problems in other fields. The CSGNN code and datasets are publicly available at https://github.com/xxhu94/CSGNN.
[ { "version": "v1", "created": "Tue, 28 Mar 2023 01:43:32 GMT" } ]
2023-03-31T00:00:00
[ [ "Hu", "Xinxin", "" ], [ "Chen", "Haotian", "" ], [ "Chen", "Hongchang", "" ], [ "Liu", "Shuxin", "" ], [ "Li", "Xing", "" ], [ "Zhang", "Shibo", "" ], [ "Wang", "Yahui", "" ], [ "Xue", "Xiangyang", "" ] ]
TITLE: Cost Sensitive GNN-based Imbalanced Learning for Mobile Social Network Fraud Detection ABSTRACT: With the rapid development of mobile networks, the people's social contacts have been considerably facilitated. However, the rise of mobile social network fraud upon those networks, has caused a great deal of distress, in case of depleting personal and social wealth, then potentially doing significant economic harm. To detect fraudulent users, call detail record (CDR) data, which portrays the social behavior of users in mobile networks, has been widely utilized. But the imbalance problem in the aforementioned data, which could severely hinder the effectiveness of fraud detectors based on graph neural networks(GNN), has hardly been addressed in previous work. In this paper, we are going to present a novel Cost-Sensitive Graph Neural Network (CSGNN) by creatively combining cost-sensitive learning and graph neural networks. We conduct extensive experiments on two open-source realworld mobile network fraud datasets. The results show that CSGNN can effectively solve the graph imbalance problem and then achieve better detection performance than the state-of-the-art algorithms. We believe that our research can be applied to solve the graph imbalance problems in other fields. The CSGNN code and datasets are publicly available at https://github.com/xxhu94/CSGNN.
no_new_dataset
0.702313
2303.17490
Kim Sung-Bin
Kim Sung-Bin, Arda Senocak, Hyunwoo Ha, Andrew Owens, Tae-Hyun Oh
Sound to Visual Scene Generation by Audio-to-Visual Latent Alignment
CVPR 2023
null
null
null
cs.CV cs.MM cs.SD eess.AS eess.IV
http://creativecommons.org/licenses/by/4.0/
How does audio describe the world around us? In this paper, we propose a method for generating an image of a scene from sound. Our method addresses the challenges of dealing with the large gaps that often exist between sight and sound. We design a model that works by scheduling the learning procedure of each model component to associate audio-visual modalities despite their information gaps. The key idea is to enrich the audio features with visual information by learning to align audio to visual latent space. We translate the input audio to visual features, then use a pre-trained generator to produce an image. To further improve the quality of our generated images, we use sound source localization to select the audio-visual pairs that have strong cross-modal correlations. We obtain substantially better results on the VEGAS and VGGSound datasets than prior approaches. We also show that we can control our model's predictions by applying simple manipulations to the input waveform, or to the latent space.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 16:01:50 GMT" } ]
2023-03-31T00:00:00
[ [ "Sung-Bin", "Kim", "" ], [ "Senocak", "Arda", "" ], [ "Ha", "Hyunwoo", "" ], [ "Owens", "Andrew", "" ], [ "Oh", "Tae-Hyun", "" ] ]
TITLE: Sound to Visual Scene Generation by Audio-to-Visual Latent Alignment ABSTRACT: How does audio describe the world around us? In this paper, we propose a method for generating an image of a scene from sound. Our method addresses the challenges of dealing with the large gaps that often exist between sight and sound. We design a model that works by scheduling the learning procedure of each model component to associate audio-visual modalities despite their information gaps. The key idea is to enrich the audio features with visual information by learning to align audio to visual latent space. We translate the input audio to visual features, then use a pre-trained generator to produce an image. To further improve the quality of our generated images, we use sound source localization to select the audio-visual pairs that have strong cross-modal correlations. We obtain substantially better results on the VEGAS and VGGSound datasets than prior approaches. We also show that we can control our model's predictions by applying simple manipulations to the input waveform, or to the latent space.
no_new_dataset
0.710597
2303.17505
Mingqing Wang
Mingqing Wang and Jiawei Li and Zhenyang Li and Chengxiao Luo and Bin Chen and Shu-Tao Xia and Zhi Wang
Unsupervised Anomaly Detection with Local-Sensitive VQVAE and Global-Sensitive Transformers
4 pages, 3 figures, 4 tables
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Unsupervised anomaly detection (UAD) has been widely implemented in industrial and medical applications, which reduces the cost of manual annotation and improves efficiency in disease diagnosis. Recently, deep auto-encoder with its variants has demonstrated its advantages in many UAD scenarios. Training on the normal data, these models are expected to locate anomalies by producing higher reconstruction error for the abnormal areas than the normal ones. However, this assumption does not always hold because of the uncontrollable generalization capability. To solve this problem, we present LSGS, a method that builds on Vector Quantised-Variational Autoencoder (VQVAE) with a novel aggregated codebook and transformers with global attention. In this work, the VQVAE focus on feature extraction and reconstruction of images, and the transformers fit the manifold and locate anomalies in the latent space. Then, leveraging the generated encoding sequences that conform to a normal distribution, we can reconstruct a more accurate image for locating the anomalies. Experiments on various datasets demonstrate the effectiveness of the proposed method.
[ { "version": "v1", "created": "Wed, 29 Mar 2023 09:13:20 GMT" } ]
2023-03-31T00:00:00
[ [ "Wang", "Mingqing", "" ], [ "Li", "Jiawei", "" ], [ "Li", "Zhenyang", "" ], [ "Luo", "Chengxiao", "" ], [ "Chen", "Bin", "" ], [ "Xia", "Shu-Tao", "" ], [ "Wang", "Zhi", "" ] ]
TITLE: Unsupervised Anomaly Detection with Local-Sensitive VQVAE and Global-Sensitive Transformers ABSTRACT: Unsupervised anomaly detection (UAD) has been widely implemented in industrial and medical applications, which reduces the cost of manual annotation and improves efficiency in disease diagnosis. Recently, deep auto-encoder with its variants has demonstrated its advantages in many UAD scenarios. Training on the normal data, these models are expected to locate anomalies by producing higher reconstruction error for the abnormal areas than the normal ones. However, this assumption does not always hold because of the uncontrollable generalization capability. To solve this problem, we present LSGS, a method that builds on Vector Quantised-Variational Autoencoder (VQVAE) with a novel aggregated codebook and transformers with global attention. In this work, the VQVAE focus on feature extraction and reconstruction of images, and the transformers fit the manifold and locate anomalies in the latent space. Then, leveraging the generated encoding sequences that conform to a normal distribution, we can reconstruct a more accurate image for locating the anomalies. Experiments on various datasets demonstrate the effectiveness of the proposed method.
no_new_dataset
0.709069
2303.17512
Georgios Chondrogiannis
Georgios D. Chondrogiannis, Nikos A. Mitsiou, Nestor D. Chatzidiamantis, Alexandros-Apostolos A. Boulogeorgos, George K. Karagiannidis
Power-Optimal HARQ Protocol for Reliable Free Space Optical Communication
6 pages, 4 figures, IEEE International Conference on Communications (ICC) 2023 WS-19: 5th Workshop on ULMC6GN - Ultra-high speed, Low latency and Massive Communication for futuristic 6G Networks
null
null
null
cs.IT eess.SP math.IT
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper investigates the usage of hybrid automatic repeat request (HARQ) protocols for power-efficient and reliable communications over free space optical (FSO) links. By exploiting the large coherence time of the FSO channel, the proposed transmission schemes combat turbulence-induced fading by retransmitting the failed packets in the same coherence interval. To assess the performance of the presented HARQ technique, we extract a theoretical framework for the outage performance. In more detail, a closed-form expression for the outage probability (OP) is reported and an approximation for the high signal-to-noise ratio (SNR) region is extracted. Building upon the theoretical framework, we formulate a transmission power allocation problem throughout the retransmission rounds. This optimization problem is solved numerically through the use of an iterative algorithm. In addition, the average throughput of the HARQ schemes under consideration is examined. Simulation results validate the theoretical analysis under different turbulence conditions and demonstrate the performance improvement, in terms of both OP and throughput, of the proposed HARQ schemes compared to fixed transmit power HARQ benchmarks.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 16:27:06 GMT" } ]
2023-03-31T00:00:00
[ [ "Chondrogiannis", "Georgios D.", "" ], [ "Mitsiou", "Nikos A.", "" ], [ "Chatzidiamantis", "Nestor D.", "" ], [ "Boulogeorgos", "Alexandros-Apostolos A.", "" ], [ "Karagiannidis", "George K.", "" ] ]
TITLE: Power-Optimal HARQ Protocol for Reliable Free Space Optical Communication ABSTRACT: This paper investigates the usage of hybrid automatic repeat request (HARQ) protocols for power-efficient and reliable communications over free space optical (FSO) links. By exploiting the large coherence time of the FSO channel, the proposed transmission schemes combat turbulence-induced fading by retransmitting the failed packets in the same coherence interval. To assess the performance of the presented HARQ technique, we extract a theoretical framework for the outage performance. In more detail, a closed-form expression for the outage probability (OP) is reported and an approximation for the high signal-to-noise ratio (SNR) region is extracted. Building upon the theoretical framework, we formulate a transmission power allocation problem throughout the retransmission rounds. This optimization problem is solved numerically through the use of an iterative algorithm. In addition, the average throughput of the HARQ schemes under consideration is examined. Simulation results validate the theoretical analysis under different turbulence conditions and demonstrate the performance improvement, in terms of both OP and throughput, of the proposed HARQ schemes compared to fixed transmit power HARQ benchmarks.
no_new_dataset
0.710201
2303.17514
Zishuo Li
Zishuo Li, Anh Tung Nguyen, Andr\'e Teixeira, Yilin Mo, Karl H. Johansson
Secure State Estimation with Asynchronous Measurements against Malicious Measurement-data and Time-stamp Manipulation
null
null
null
null
eess.SY cs.SY
http://creativecommons.org/licenses/by-nc-nd/4.0/
This paper proposes a secure state estimation scheme with non-periodic asynchronous measurements for linear continuous-time systems under false data attacks on the measurement transmit channel. After sampling the output of the system, a sensor transmits the measurement information in a triple composed of sensor index, time-stamp, and measurement value to the fusion center via vulnerable communication channels. The malicious attacker can corrupt a subset of the sensors through (i) manipulating the time-stamp and measurement value; (ii) blocking transmitted measurement triples; or (iii) injecting fake measurement triples. To deal with such attacks, we propose the design of local estimators based on observability space decomposition, where each local estimator updates the local state and sends it to the fusion center after sampling a measurement. Whenever there is a local update, the fusion center combines all the local states and generates a secure state estimate by adopting the median operator. We prove that local estimators of benign sensors are unbiased with stable covariance. Moreover, the fused central estimation error has bounded expectation and covariance against at most $p$ corrupted sensors as long as the system is $2p$-sparse observable. The efficacy of the proposed scheme is demonstrated through an application on a benchmark example of the IEEE 14-bus system.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 16:28:36 GMT" } ]
2023-03-31T00:00:00
[ [ "Li", "Zishuo", "" ], [ "Nguyen", "Anh Tung", "" ], [ "Teixeira", "André", "" ], [ "Mo", "Yilin", "" ], [ "Johansson", "Karl H.", "" ] ]
TITLE: Secure State Estimation with Asynchronous Measurements against Malicious Measurement-data and Time-stamp Manipulation ABSTRACT: This paper proposes a secure state estimation scheme with non-periodic asynchronous measurements for linear continuous-time systems under false data attacks on the measurement transmit channel. After sampling the output of the system, a sensor transmits the measurement information in a triple composed of sensor index, time-stamp, and measurement value to the fusion center via vulnerable communication channels. The malicious attacker can corrupt a subset of the sensors through (i) manipulating the time-stamp and measurement value; (ii) blocking transmitted measurement triples; or (iii) injecting fake measurement triples. To deal with such attacks, we propose the design of local estimators based on observability space decomposition, where each local estimator updates the local state and sends it to the fusion center after sampling a measurement. Whenever there is a local update, the fusion center combines all the local states and generates a secure state estimate by adopting the median operator. We prove that local estimators of benign sensors are unbiased with stable covariance. Moreover, the fused central estimation error has bounded expectation and covariance against at most $p$ corrupted sensors as long as the system is $2p$-sparse observable. The efficacy of the proposed scheme is demonstrated through an application on a benchmark example of the IEEE 14-bus system.
no_new_dataset
0.708395
2303.17526
Wenqiao Zhang
Wenqiao Zhang, Changshuo Liu, Can Cui, Beng Chin Ooi
CAusal and collaborative proxy-tasKs lEarning for Semi-Supervised Domain Adaptation
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Semi-supervised domain adaptation (SSDA) adapts a learner to a new domain by effectively utilizing source domain data and a few labeled target samples. It is a practical yet under-investigated research topic. In this paper, we analyze the SSDA problem from two perspectives that have previously been overlooked, and correspondingly decompose it into two \emph{key subproblems}: \emph{robust domain adaptation (DA) learning} and \emph{maximal cross-domain data utilization}. \textbf{(i)} From a causal theoretical view, a robust DA model should distinguish the invariant ``concept'' (key clue to image label) from the nuisance of confounding factors across domains. To achieve this goal, we propose to generate \emph{concept-invariant samples} to enable the model to classify the samples through causal intervention, yielding improved generalization guarantees; \textbf{(ii)} Based on the robust DA theory, we aim to exploit the maximal utilization of rich source domain data and a few labeled target samples to boost SSDA further. Consequently, we propose a collaboratively debiasing learning framework that utilizes two complementary semi-supervised learning (SSL) classifiers to mutually exchange their unbiased knowledge, which helps unleash the potential of source and target domain training data, thereby producing more convincing pseudo-labels. Such obtained labels facilitate cross-domain feature alignment and duly improve the invariant concept learning. In our experimental study, we show that the proposed model significantly outperforms SOTA methods in terms of effectiveness and generalisability on SSDA datasets.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 16:48:28 GMT" } ]
2023-03-31T00:00:00
[ [ "Zhang", "Wenqiao", "" ], [ "Liu", "Changshuo", "" ], [ "Cui", "Can", "" ], [ "Ooi", "Beng Chin", "" ] ]
TITLE: CAusal and collaborative proxy-tasKs lEarning for Semi-Supervised Domain Adaptation ABSTRACT: Semi-supervised domain adaptation (SSDA) adapts a learner to a new domain by effectively utilizing source domain data and a few labeled target samples. It is a practical yet under-investigated research topic. In this paper, we analyze the SSDA problem from two perspectives that have previously been overlooked, and correspondingly decompose it into two \emph{key subproblems}: \emph{robust domain adaptation (DA) learning} and \emph{maximal cross-domain data utilization}. \textbf{(i)} From a causal theoretical view, a robust DA model should distinguish the invariant ``concept'' (key clue to image label) from the nuisance of confounding factors across domains. To achieve this goal, we propose to generate \emph{concept-invariant samples} to enable the model to classify the samples through causal intervention, yielding improved generalization guarantees; \textbf{(ii)} Based on the robust DA theory, we aim to exploit the maximal utilization of rich source domain data and a few labeled target samples to boost SSDA further. Consequently, we propose a collaboratively debiasing learning framework that utilizes two complementary semi-supervised learning (SSL) classifiers to mutually exchange their unbiased knowledge, which helps unleash the potential of source and target domain training data, thereby producing more convincing pseudo-labels. Such obtained labels facilitate cross-domain feature alignment and duly improve the invariant concept learning. In our experimental study, we show that the proposed model significantly outperforms SOTA methods in terms of effectiveness and generalisability on SSDA datasets.
no_new_dataset
0.710025
2303.17548
Shibani Santurkar
Shibani Santurkar, Esin Durmus, Faisal Ladhak, Cinoo Lee, Percy Liang, Tatsunori Hashimoto
Whose Opinions Do Language Models Reflect?
null
null
null
null
cs.CL cs.AI cs.CY cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Language models (LMs) are increasingly being used in open-ended contexts, where the opinions reflected by LMs in response to subjective queries can have a profound impact, both on user satisfaction, as well as shaping the views of society at large. In this work, we put forth a quantitative framework to investigate the opinions reflected by LMs -- by leveraging high-quality public opinion polls and their associated human responses. Using this framework, we create OpinionsQA, a new dataset for evaluating the alignment of LM opinions with those of 60 US demographic groups over topics ranging from abortion to automation. Across topics, we find substantial misalignment between the views reflected by current LMs and those of US demographic groups: on par with the Democrat-Republican divide on climate change. Notably, this misalignment persists even after explicitly steering the LMs towards particular demographic groups. Our analysis not only confirms prior observations about the left-leaning tendencies of some human feedback-tuned LMs, but also surfaces groups whose opinions are poorly reflected by current LMs (e.g., 65+ and widowed individuals). Our code and data are available at https://github.com/tatsu-lab/opinions_qa.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 17:17:08 GMT" } ]
2023-03-31T00:00:00
[ [ "Santurkar", "Shibani", "" ], [ "Durmus", "Esin", "" ], [ "Ladhak", "Faisal", "" ], [ "Lee", "Cinoo", "" ], [ "Liang", "Percy", "" ], [ "Hashimoto", "Tatsunori", "" ] ]
TITLE: Whose Opinions Do Language Models Reflect? ABSTRACT: Language models (LMs) are increasingly being used in open-ended contexts, where the opinions reflected by LMs in response to subjective queries can have a profound impact, both on user satisfaction, as well as shaping the views of society at large. In this work, we put forth a quantitative framework to investigate the opinions reflected by LMs -- by leveraging high-quality public opinion polls and their associated human responses. Using this framework, we create OpinionsQA, a new dataset for evaluating the alignment of LM opinions with those of 60 US demographic groups over topics ranging from abortion to automation. Across topics, we find substantial misalignment between the views reflected by current LMs and those of US demographic groups: on par with the Democrat-Republican divide on climate change. Notably, this misalignment persists even after explicitly steering the LMs towards particular demographic groups. Our analysis not only confirms prior observations about the left-leaning tendencies of some human feedback-tuned LMs, but also surfaces groups whose opinions are poorly reflected by current LMs (e.g., 65+ and widowed individuals). Our code and data are available at https://github.com/tatsu-lab/opinions_qa.
new_dataset
0.713475
2303.17574
Nico Daheim
Nico Daheim, Nouha Dziri, Mrinmaya Sachan, Iryna Gurevych, Edoardo M. Ponti
Elastic Weight Removal for Faithful and Abstractive Dialogue Generation
null
null
null
null
cs.CL cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
Ideally, dialogue systems should generate responses that are faithful to the knowledge contained in relevant documents. However, many models generate hallucinated responses instead that contradict it or contain unverifiable information. To mitigate such undesirable behaviour, it has been proposed to fine-tune a `negative expert' on negative examples and subtract its parameters from those of a pre-trained model. However, intuitively, this does not take into account that some parameters are more responsible than others in causing hallucinations. Thus, we propose to weigh their individual importance via (an approximation of) the Fisher Information matrix, which measures the uncertainty of their estimate. We call this method Elastic Weight Removal (EWR). We evaluate our method -- using different variants of Flan-T5 as a backbone language model -- on multiple datasets for information-seeking dialogue generation and compare our method with state-of-the-art techniques for faithfulness, such as CTRL, Quark, DExperts, and Noisy Channel reranking. Extensive automatic and human evaluation shows that EWR systematically increases faithfulness at minor costs in terms of other metrics. However, we notice that only discouraging hallucinations may increase extractiveness, i.e. shallow copy-pasting of document spans, which can be undesirable. Hence, as a second main contribution, we show that our method can be extended to simultaneously discourage hallucinations and extractive responses. We publicly release the code for reproducing EWR and all baselines.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 17:40:30 GMT" } ]
2023-03-31T00:00:00
[ [ "Daheim", "Nico", "" ], [ "Dziri", "Nouha", "" ], [ "Sachan", "Mrinmaya", "" ], [ "Gurevych", "Iryna", "" ], [ "Ponti", "Edoardo M.", "" ] ]
TITLE: Elastic Weight Removal for Faithful and Abstractive Dialogue Generation ABSTRACT: Ideally, dialogue systems should generate responses that are faithful to the knowledge contained in relevant documents. However, many models generate hallucinated responses instead that contradict it or contain unverifiable information. To mitigate such undesirable behaviour, it has been proposed to fine-tune a `negative expert' on negative examples and subtract its parameters from those of a pre-trained model. However, intuitively, this does not take into account that some parameters are more responsible than others in causing hallucinations. Thus, we propose to weigh their individual importance via (an approximation of) the Fisher Information matrix, which measures the uncertainty of their estimate. We call this method Elastic Weight Removal (EWR). We evaluate our method -- using different variants of Flan-T5 as a backbone language model -- on multiple datasets for information-seeking dialogue generation and compare our method with state-of-the-art techniques for faithfulness, such as CTRL, Quark, DExperts, and Noisy Channel reranking. Extensive automatic and human evaluation shows that EWR systematically increases faithfulness at minor costs in terms of other metrics. However, we notice that only discouraging hallucinations may increase extractiveness, i.e. shallow copy-pasting of document spans, which can be undesirable. Hence, as a second main contribution, we show that our method can be extended to simultaneously discourage hallucinations and extractive responses. We publicly release the code for reproducing EWR and all baselines.
no_new_dataset
0.709038
2303.17592
Sammy Christen
Sammy Christen, Wei Yang, Claudia P\'erez-D'Arpino, Otmar Hilliges, Dieter Fox, Yu-Wei Chao
Learning Human-to-Robot Handovers from Point Clouds
Accepted at CVPR 2023 as highlight. Project page at https://handover-sim2real.github.io
null
null
null
cs.RO cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We propose the first framework to learn control policies for vision-based human-to-robot handovers, a critical task for human-robot interaction. While research in Embodied AI has made significant progress in training robot agents in simulated environments, interacting with humans remains challenging due to the difficulties of simulating humans. Fortunately, recent research has developed realistic simulated environments for human-to-robot handovers. Leveraging this result, we introduce a method that is trained with a human-in-the-loop via a two-stage teacher-student framework that uses motion and grasp planning, reinforcement learning, and self-supervision. We show significant performance gains over baselines on a simulation benchmark, sim-to-sim transfer and sim-to-real transfer.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 17:58:36 GMT" } ]
2023-03-31T00:00:00
[ [ "Christen", "Sammy", "" ], [ "Yang", "Wei", "" ], [ "Pérez-D'Arpino", "Claudia", "" ], [ "Hilliges", "Otmar", "" ], [ "Fox", "Dieter", "" ], [ "Chao", "Yu-Wei", "" ] ]
TITLE: Learning Human-to-Robot Handovers from Point Clouds ABSTRACT: We propose the first framework to learn control policies for vision-based human-to-robot handovers, a critical task for human-robot interaction. While research in Embodied AI has made significant progress in training robot agents in simulated environments, interacting with humans remains challenging due to the difficulties of simulating humans. Fortunately, recent research has developed realistic simulated environments for human-to-robot handovers. Leveraging this result, we introduce a method that is trained with a human-in-the-loop via a two-stage teacher-student framework that uses motion and grasp planning, reinforcement learning, and self-supervision. We show significant performance gains over baselines on a simulation benchmark, sim-to-sim transfer and sim-to-real transfer.
no_new_dataset
0.709466
2303.17598
Hung-Yu Tseng
Hung-Yu Tseng, Qinbo Li, Changil Kim, Suhib Alsisan, Jia-Bin Huang, Johannes Kopf
Consistent View Synthesis with Pose-Guided Diffusion Models
CVPR 2023. Project page: https://poseguided-diffusion.github.io/
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Novel view synthesis from a single image has been a cornerstone problem for many Virtual Reality applications that provide immersive experiences. However, most existing techniques can only synthesize novel views within a limited range of camera motion or fail to generate consistent and high-quality novel views under significant camera movement. In this work, we propose a pose-guided diffusion model to generate a consistent long-term video of novel views from a single image. We design an attention layer that uses epipolar lines as constraints to facilitate the association between different viewpoints. Experimental results on synthetic and real-world datasets demonstrate the effectiveness of the proposed diffusion model against state-of-the-art transformer-based and GAN-based approaches.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 17:59:22 GMT" } ]
2023-03-31T00:00:00
[ [ "Tseng", "Hung-Yu", "" ], [ "Li", "Qinbo", "" ], [ "Kim", "Changil", "" ], [ "Alsisan", "Suhib", "" ], [ "Huang", "Jia-Bin", "" ], [ "Kopf", "Johannes", "" ] ]
TITLE: Consistent View Synthesis with Pose-Guided Diffusion Models ABSTRACT: Novel view synthesis from a single image has been a cornerstone problem for many Virtual Reality applications that provide immersive experiences. However, most existing techniques can only synthesize novel views within a limited range of camera motion or fail to generate consistent and high-quality novel views under significant camera movement. In this work, we propose a pose-guided diffusion model to generate a consistent long-term video of novel views from a single image. We design an attention layer that uses epipolar lines as constraints to facilitate the association between different viewpoints. Experimental results on synthetic and real-world datasets demonstrate the effectiveness of the proposed diffusion model against state-of-the-art transformer-based and GAN-based approaches.
no_new_dataset
0.712251
2303.17600
Zichen Zhang
Zichen Zhang, Luca Weihs
When Learning Is Out of Reach, Reset: Generalization in Autonomous Visuomotor Reinforcement Learning
null
null
null
null
cs.CV cs.RO
http://creativecommons.org/licenses/by-nc-sa/4.0/
Episodic training, where an agent's environment is reset after every success or failure, is the de facto standard when training embodied reinforcement learning (RL) agents. The underlying assumption that the environment can be easily reset is limiting both practically, as resets generally require human effort in the real world and can be computationally expensive in simulation, and philosophically, as we'd expect intelligent agents to be able to continuously learn without intervention. Work in learning without any resets, i.e{.} Reset-Free RL (RF-RL), is promising but is plagued by the problem of irreversible transitions (e.g{.} an object breaking) which halt learning. Moreover, the limited state diversity and instrument setup encountered during RF-RL means that works studying RF-RL largely do not require their models to generalize to new environments. In this work, we instead look to minimize, rather than completely eliminate, resets while building visual agents that can meaningfully generalize. As studying generalization has previously not been a focus of benchmarks designed for RF-RL, we propose a new Stretch Pick-and-Place benchmark designed for evaluating generalizations across goals, cosmetic variations, and structural changes. Moreover, towards building performant reset-minimizing RL agents, we propose unsupervised metrics to detect irreversible transitions and a single-policy training mechanism to enable generalization. Our proposed approach significantly outperforms prior episodic, reset-free, and reset-minimizing approaches achieving higher success rates with fewer resets in Stretch-P\&P and another popular RF-RL benchmark. Finally, we find that our proposed approach can dramatically reduce the number of resets required for training other embodied tasks, in particular for RoboTHOR ObjectNav we obtain higher success rates than episodic approaches using 99.97\% fewer resets.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 17:59:26 GMT" } ]
2023-03-31T00:00:00
[ [ "Zhang", "Zichen", "" ], [ "Weihs", "Luca", "" ] ]
TITLE: When Learning Is Out of Reach, Reset: Generalization in Autonomous Visuomotor Reinforcement Learning ABSTRACT: Episodic training, where an agent's environment is reset after every success or failure, is the de facto standard when training embodied reinforcement learning (RL) agents. The underlying assumption that the environment can be easily reset is limiting both practically, as resets generally require human effort in the real world and can be computationally expensive in simulation, and philosophically, as we'd expect intelligent agents to be able to continuously learn without intervention. Work in learning without any resets, i.e{.} Reset-Free RL (RF-RL), is promising but is plagued by the problem of irreversible transitions (e.g{.} an object breaking) which halt learning. Moreover, the limited state diversity and instrument setup encountered during RF-RL means that works studying RF-RL largely do not require their models to generalize to new environments. In this work, we instead look to minimize, rather than completely eliminate, resets while building visual agents that can meaningfully generalize. As studying generalization has previously not been a focus of benchmarks designed for RF-RL, we propose a new Stretch Pick-and-Place benchmark designed for evaluating generalizations across goals, cosmetic variations, and structural changes. Moreover, towards building performant reset-minimizing RL agents, we propose unsupervised metrics to detect irreversible transitions and a single-policy training mechanism to enable generalization. Our proposed approach significantly outperforms prior episodic, reset-free, and reset-minimizing approaches achieving higher success rates with fewer resets in Stretch-P\&P and another popular RF-RL benchmark. Finally, we find that our proposed approach can dramatically reduce the number of resets required for training other embodied tasks, in particular for RoboTHOR ObjectNav we obtain higher success rates than episodic approaches using 99.97\% fewer resets.
no_new_dataset
0.710226
2303.17603
Matteo Poggi
Fabio Tosi, Alessio Tonioni, Daniele De Gregorio, Matteo Poggi
NeRF-Supervised Deep Stereo
CVPR 2023. Project page: https://nerfstereo.github.io/ Code: https://github.com/fabiotosi92/NeRF-Supervised-Deep-Stereo
null
null
null
cs.CV cs.RO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We introduce a novel framework for training deep stereo networks effortlessly and without any ground-truth. By leveraging state-of-the-art neural rendering solutions, we generate stereo training data from image sequences collected with a single handheld camera. On top of them, a NeRF-supervised training procedure is carried out, from which we exploit rendered stereo triplets to compensate for occlusions and depth maps as proxy labels. This results in stereo networks capable of predicting sharp and detailed disparity maps. Experimental results show that models trained under this regime yield a 30-40% improvement over existing self-supervised methods on the challenging Middlebury dataset, filling the gap to supervised models and, most times, outperforming them at zero-shot generalization.
[ { "version": "v1", "created": "Thu, 30 Mar 2023 17:59:58 GMT" } ]
2023-03-31T00:00:00
[ [ "Tosi", "Fabio", "" ], [ "Tonioni", "Alessio", "" ], [ "De Gregorio", "Daniele", "" ], [ "Poggi", "Matteo", "" ] ]
TITLE: NeRF-Supervised Deep Stereo ABSTRACT: We introduce a novel framework for training deep stereo networks effortlessly and without any ground-truth. By leveraging state-of-the-art neural rendering solutions, we generate stereo training data from image sequences collected with a single handheld camera. On top of them, a NeRF-supervised training procedure is carried out, from which we exploit rendered stereo triplets to compensate for occlusions and depth maps as proxy labels. This results in stereo networks capable of predicting sharp and detailed disparity maps. Experimental results show that models trained under this regime yield a 30-40% improvement over existing self-supervised methods on the challenging Middlebury dataset, filling the gap to supervised models and, most times, outperforming them at zero-shot generalization.
no_new_dataset
0.710239
2106.10365
Abdus Salam Azad
Abdus Salam Azad, Edward Kim, Qiancheng Wu, Kimin Lee, Ion Stoica, Pieter Abbeel, and Sanjit A. Seshia
Scenic4RL: Programmatic Modeling and Generation of Reinforcement Learning Environments
First two authors contributed equally. The final version of this paper is accepted at Proceedings of the AAAI Conference on Artificial Intelligence, 36(6), 6028-6036. https://doi.org/10.1609/aaai.v36i6.20549
null
null
null
cs.LG cs.AI
http://creativecommons.org/licenses/by/4.0/
The capability of a reinforcement learning (RL) agent heavily depends on the diversity of the learning scenarios generated by the environment. Generation of diverse realistic scenarios is challenging for real-time strategy (RTS) environments. The RTS environments are characterized by intelligent entities/non-RL agents cooperating and competing with the RL agents with large state and action spaces over a long period of time, resulting in an infinite space of feasible, but not necessarily realistic, scenarios involving complex interaction among different RL and non-RL agents. Yet, most of the existing simulators rely on randomly generating the environments based on predefined settings/layouts and offer limited flexibility and control over the environment dynamics for researchers to generate diverse, realistic scenarios as per their demand. To address this issue, for the first time, we formally introduce the benefits of adopting an existing formal scenario specification language, SCENIC, to assist researchers to model and generate diverse scenarios in an RTS environment in a flexible, systematic, and programmatic manner. To showcase the benefits, we interfaced SCENIC to an existing RTS environment Google Research Football(GRF) simulator and introduced a benchmark consisting of 32 realistic scenarios, encoded in SCENIC, to train RL agents and testing their generalization capabilities. We also show how researchers/RL practitioners can incorporate their domain knowledge to expedite the training process by intuitively modeling stochastic programmatic policies with SCENIC.
[ { "version": "v1", "created": "Fri, 18 Jun 2021 21:49:46 GMT" }, { "version": "v2", "created": "Tue, 28 Mar 2023 20:25:15 GMT" } ]
2023-03-30T00:00:00
[ [ "Azad", "Abdus Salam", "" ], [ "Kim", "Edward", "" ], [ "Wu", "Qiancheng", "" ], [ "Lee", "Kimin", "" ], [ "Stoica", "Ion", "" ], [ "Abbeel", "Pieter", "" ], [ "Seshia", "Sanjit A.", "" ] ]
TITLE: Scenic4RL: Programmatic Modeling and Generation of Reinforcement Learning Environments ABSTRACT: The capability of a reinforcement learning (RL) agent heavily depends on the diversity of the learning scenarios generated by the environment. Generation of diverse realistic scenarios is challenging for real-time strategy (RTS) environments. The RTS environments are characterized by intelligent entities/non-RL agents cooperating and competing with the RL agents with large state and action spaces over a long period of time, resulting in an infinite space of feasible, but not necessarily realistic, scenarios involving complex interaction among different RL and non-RL agents. Yet, most of the existing simulators rely on randomly generating the environments based on predefined settings/layouts and offer limited flexibility and control over the environment dynamics for researchers to generate diverse, realistic scenarios as per their demand. To address this issue, for the first time, we formally introduce the benefits of adopting an existing formal scenario specification language, SCENIC, to assist researchers to model and generate diverse scenarios in an RTS environment in a flexible, systematic, and programmatic manner. To showcase the benefits, we interfaced SCENIC to an existing RTS environment Google Research Football(GRF) simulator and introduced a benchmark consisting of 32 realistic scenarios, encoded in SCENIC, to train RL agents and testing their generalization capabilities. We also show how researchers/RL practitioners can incorporate their domain knowledge to expedite the training process by intuitively modeling stochastic programmatic policies with SCENIC.
no_new_dataset
0.703384
2106.12096
Marissa Connor
Marissa Connor, Kion Fallah, Christopher Rozell
Learning Identity-Preserving Transformations on Data Manifolds
null
null
null
null
cs.LG stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Many machine learning techniques incorporate identity-preserving transformations into their models to generalize their performance to previously unseen data. These transformations are typically selected from a set of functions that are known to maintain the identity of an input when applied (e.g., rotation, translation, flipping, and scaling). However, there are many natural variations that cannot be labeled for supervision or defined through examination of the data. As suggested by the manifold hypothesis, many of these natural variations live on or near a low-dimensional, nonlinear manifold. Several techniques represent manifold variations through a set of learned Lie group operators that define directions of motion on the manifold. However, these approaches are limited because they require transformation labels when training their models and they lack a method for determining which regions of the manifold are appropriate for applying each specific operator. We address these limitations by introducing a learning strategy that does not require transformation labels and developing a method that learns the local regions where each operator is likely to be used while preserving the identity of inputs. Experiments on MNIST and Fashion MNIST highlight our model's ability to learn identity-preserving transformations on multi-class datasets. Additionally, we train on CelebA to showcase our model's ability to learn semantically meaningful transformations on complex datasets in an unsupervised manner.
[ { "version": "v1", "created": "Tue, 22 Jun 2021 23:10:25 GMT" }, { "version": "v2", "created": "Wed, 29 Mar 2023 03:12:54 GMT" } ]
2023-03-30T00:00:00
[ [ "Connor", "Marissa", "" ], [ "Fallah", "Kion", "" ], [ "Rozell", "Christopher", "" ] ]
TITLE: Learning Identity-Preserving Transformations on Data Manifolds ABSTRACT: Many machine learning techniques incorporate identity-preserving transformations into their models to generalize their performance to previously unseen data. These transformations are typically selected from a set of functions that are known to maintain the identity of an input when applied (e.g., rotation, translation, flipping, and scaling). However, there are many natural variations that cannot be labeled for supervision or defined through examination of the data. As suggested by the manifold hypothesis, many of these natural variations live on or near a low-dimensional, nonlinear manifold. Several techniques represent manifold variations through a set of learned Lie group operators that define directions of motion on the manifold. However, these approaches are limited because they require transformation labels when training their models and they lack a method for determining which regions of the manifold are appropriate for applying each specific operator. We address these limitations by introducing a learning strategy that does not require transformation labels and developing a method that learns the local regions where each operator is likely to be used while preserving the identity of inputs. Experiments on MNIST and Fashion MNIST highlight our model's ability to learn identity-preserving transformations on multi-class datasets. Additionally, we train on CelebA to showcase our model's ability to learn semantically meaningful transformations on complex datasets in an unsupervised manner.
no_new_dataset
0.709202
2110.07557
Zhemin Li
Zhemin Li, Tao Sun, Hongxia Wang, Bao Wang
AIR-Net: Adaptive and Implicit Regularization Neural Network for Matrix Completion
null
SIAM Journal on Imaging Sciences (2022)
10.1137/22M1489228
null
cs.LG stat.ML
http://creativecommons.org/licenses/by/4.0/
The explicit low-rank regularization, e.g., nuclear norm regularization, has been widely used in imaging sciences. However, it has been found that implicit regularization outperforms explicit ones in various image processing tasks. Another issue is that the fixed explicit regularization limits the applicability to broad images since different images favor different features captured by different explicit regularizations. As such, this paper proposes a new adaptive and implicit low-rank regularization that captures the low-rank prior dynamically from the training data. The core of our new adaptive and implicit low-rank regularization is parameterizing the Laplacian matrix in the Dirichlet energy-based regularization, which we call the regularization \textit{AIR}. Theoretically, we show that the adaptive regularization of AIR enhances the implicit regularization and vanishes at the end of training. We validate AIR's effectiveness on various benchmark tasks, indicating that the AIR is particularly favorable for the scenarios when the missing entries are non-uniform. The code can be found at https://github.com/lizhemin15/AIR-Net
[ { "version": "v1", "created": "Tue, 12 Oct 2021 04:13:33 GMT" }, { "version": "v2", "created": "Wed, 29 Dec 2021 01:12:34 GMT" }, { "version": "v3", "created": "Thu, 6 Jan 2022 08:54:30 GMT" }, { "version": "v4", "created": "Mon, 27 Jun 2022 01:26:22 GMT" } ]
2023-03-30T00:00:00
[ [ "Li", "Zhemin", "" ], [ "Sun", "Tao", "" ], [ "Wang", "Hongxia", "" ], [ "Wang", "Bao", "" ] ]
TITLE: AIR-Net: Adaptive and Implicit Regularization Neural Network for Matrix Completion ABSTRACT: The explicit low-rank regularization, e.g., nuclear norm regularization, has been widely used in imaging sciences. However, it has been found that implicit regularization outperforms explicit ones in various image processing tasks. Another issue is that the fixed explicit regularization limits the applicability to broad images since different images favor different features captured by different explicit regularizations. As such, this paper proposes a new adaptive and implicit low-rank regularization that captures the low-rank prior dynamically from the training data. The core of our new adaptive and implicit low-rank regularization is parameterizing the Laplacian matrix in the Dirichlet energy-based regularization, which we call the regularization \textit{AIR}. Theoretically, we show that the adaptive regularization of AIR enhances the implicit regularization and vanishes at the end of training. We validate AIR's effectiveness on various benchmark tasks, indicating that the AIR is particularly favorable for the scenarios when the missing entries are non-uniform. The code can be found at https://github.com/lizhemin15/AIR-Net
no_new_dataset
0.709372
2111.09537
MD Abdullah Al Nasim
MD Abdullah Al Nasim, Aditi Dhali, Faria Afrin, Noshin Tasnim Zaman, Nazmul Karimm, Md Mahim Anjum Haque
The Prominence of Artificial Intelligence in COVID-19
63 pages, 3 tables, 17 figures
null
null
null
cs.LG cs.AI
http://creativecommons.org/publicdomain/zero/1.0/
In December 2019, a novel virus called COVID-19 had caused an enormous number of causalities to date. The battle with the novel Coronavirus is baffling and horrifying after the Spanish Flu 2019. While the front-line doctors and medical researchers have made significant progress in controlling the spread of the highly contiguous virus, technology has also proved its significance in the battle. Moreover, Artificial Intelligence has been adopted in many medical applications to diagnose many diseases, even baffling experienced doctors. Therefore, this survey paper explores the methodologies proposed that can aid doctors and researchers in early and inexpensive methods of diagnosis of the disease. Most developing countries have difficulties carrying out tests using the conventional manner, but a significant way can be adopted with Machine and Deep Learning. On the other hand, the access to different types of medical images has motivated the researchers. As a result, a mammoth number of techniques are proposed. This paper first details the background knowledge of the conventional methods in the Artificial Intelligence domain. Following that, we gather the commonly used datasets and their use cases to date. In addition, we also show the percentage of researchers adopting Machine Learning over Deep Learning. Thus we provide a thorough analysis of this scenario. Lastly, in the research challenges, we elaborate on the problems faced in COVID-19 research, and we address the issues with our understanding to build a bright and healthy environment.
[ { "version": "v1", "created": "Thu, 18 Nov 2021 06:11:45 GMT" }, { "version": "v2", "created": "Tue, 17 Jan 2023 14:57:14 GMT" }, { "version": "v3", "created": "Wed, 29 Mar 2023 15:33:16 GMT" } ]
2023-03-30T00:00:00
[ [ "Nasim", "MD Abdullah Al", "" ], [ "Dhali", "Aditi", "" ], [ "Afrin", "Faria", "" ], [ "Zaman", "Noshin Tasnim", "" ], [ "Karimm", "Nazmul", "" ], [ "Haque", "Md Mahim Anjum", "" ] ]
TITLE: The Prominence of Artificial Intelligence in COVID-19 ABSTRACT: In December 2019, a novel virus called COVID-19 had caused an enormous number of causalities to date. The battle with the novel Coronavirus is baffling and horrifying after the Spanish Flu 2019. While the front-line doctors and medical researchers have made significant progress in controlling the spread of the highly contiguous virus, technology has also proved its significance in the battle. Moreover, Artificial Intelligence has been adopted in many medical applications to diagnose many diseases, even baffling experienced doctors. Therefore, this survey paper explores the methodologies proposed that can aid doctors and researchers in early and inexpensive methods of diagnosis of the disease. Most developing countries have difficulties carrying out tests using the conventional manner, but a significant way can be adopted with Machine and Deep Learning. On the other hand, the access to different types of medical images has motivated the researchers. As a result, a mammoth number of techniques are proposed. This paper first details the background knowledge of the conventional methods in the Artificial Intelligence domain. Following that, we gather the commonly used datasets and their use cases to date. In addition, we also show the percentage of researchers adopting Machine Learning over Deep Learning. Thus we provide a thorough analysis of this scenario. Lastly, in the research challenges, we elaborate on the problems faced in COVID-19 research, and we address the issues with our understanding to build a bright and healthy environment.
no_new_dataset
0.707234
2112.03611
Li-Hsiang Shen
Li-Hsiang Shen, Chia-Lin Tsai, Chia-Yu Wang, Kai-Ten Feng
Hybrid Controlled User Association and Resource Management for Energy-Efficient Green RANs with Limited Fronthaul
null
IEEE Access, 2022
10.1109/ACCESS.2022.3140814
null
eess.SY cs.SY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
To alleviate green house effect, high network energy efficiency (EE) has increasingly become an important research target in wireless green communications. Therefore, the investigation for resource management to mitigate the co-tier interference in the small cell network (SCN) is provided. Moreover, with the merits of cloud radio access network (C-RAN), small cell base stations (SBSs) can be decomposed of a central small cell (CSC) and remote small cells (RSCs). To achieve the coordination, the split medium access control (MAC) based functional splitting is adopted with scheduler deployed at CSCs and retransmission functions left at RSCs. However, limited fronthaul has a compelling impact at RSCs due to requirements of user quality-of-service (QoS). Accordingly, a traffic control-based user association and resource allocation (TURA) scheme is proposed for a centralized resource management. To deal with the infeasibility to control all RSCs by CSC, we propose a hybrid controlled user and resource management (HARM) scheme. A CSC performs TURA for RSCs to mitigate intra-group interference within localized C-RANs, whereas the CSCs among separate C-RANs conduct cooperative resource competition (CRC) game for alleviating inter-group interference. Based on regret-based learning algorithm, the proposed schemes are analytically proved to reach the correlated equilibrium (CE). Simulation results have validated the effect of traffic control in TURA scheme and the convergence of CRC. Moreover, the comparison of the proposed TURA, HARM, and CRC schemes with the benchmark is revealed. It is observed that the TURA scheme outperforms the other schemes under ideal fronthaul control, whilst the proposed HARM scheme can sustain EE performance considering feasible implementation.
[ { "version": "v1", "created": "Tue, 7 Dec 2021 10:13:12 GMT" } ]
2023-03-30T00:00:00
[ [ "Shen", "Li-Hsiang", "" ], [ "Tsai", "Chia-Lin", "" ], [ "Wang", "Chia-Yu", "" ], [ "Feng", "Kai-Ten", "" ] ]
TITLE: Hybrid Controlled User Association and Resource Management for Energy-Efficient Green RANs with Limited Fronthaul ABSTRACT: To alleviate green house effect, high network energy efficiency (EE) has increasingly become an important research target in wireless green communications. Therefore, the investigation for resource management to mitigate the co-tier interference in the small cell network (SCN) is provided. Moreover, with the merits of cloud radio access network (C-RAN), small cell base stations (SBSs) can be decomposed of a central small cell (CSC) and remote small cells (RSCs). To achieve the coordination, the split medium access control (MAC) based functional splitting is adopted with scheduler deployed at CSCs and retransmission functions left at RSCs. However, limited fronthaul has a compelling impact at RSCs due to requirements of user quality-of-service (QoS). Accordingly, a traffic control-based user association and resource allocation (TURA) scheme is proposed for a centralized resource management. To deal with the infeasibility to control all RSCs by CSC, we propose a hybrid controlled user and resource management (HARM) scheme. A CSC performs TURA for RSCs to mitigate intra-group interference within localized C-RANs, whereas the CSCs among separate C-RANs conduct cooperative resource competition (CRC) game for alleviating inter-group interference. Based on regret-based learning algorithm, the proposed schemes are analytically proved to reach the correlated equilibrium (CE). Simulation results have validated the effect of traffic control in TURA scheme and the convergence of CRC. Moreover, the comparison of the proposed TURA, HARM, and CRC schemes with the benchmark is revealed. It is observed that the TURA scheme outperforms the other schemes under ideal fronthaul control, whilst the proposed HARM scheme can sustain EE performance considering feasible implementation.
no_new_dataset
0.709025
2112.04070
Li-Hsiang Shen
Li-Hsiang Shen, Chia-Yu Su, Kai-Ten Feng
CoMP Enhanced Subcarrier and Power Allocation for Multi-Numerology based 5G-NR Networks
null
IEEE Transactions on Vehicular Technology, 2022
10.1109/TVT.2022.3154896
null
cs.NI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
With proliferation of fifth generation (5G) new radio (NR) technology, it is expected to meet the requirement of diverse traffic demands. We have designed a coordinated multi-point (CoMP) enhanced flexible multi-numerology (MN) for 5G-NR networks to improve the network performance in terms of throughput and latency. We have proposed a CoMP enhanced joint subcarrier and power allocation (CESP) scheme which aims at maximizing sum rate under the considerations of transmit power limitation and guaranteed quality-of-service (QoS) including throughput and latency restrictions. By employing difference of two concave functions (D.C.) approximation and abstract Lagrangian duality method, we theoretically transform the original non-convex nonlinear problem into a solvable maximization problem. Moreover, the convergence of our proposed CESP algorithm with D.C. approximation is analytically derived with proofs, and is further validated via numerical results. Simulation results demonstrated that our proposed CESP algorithm outperforms the conventional non-CoMP and single numerology mechanisms along with other existing benchmarks in terms of lower latency and higher throughput under the scenarios of uniform and edge users.
[ { "version": "v1", "created": "Wed, 8 Dec 2021 01:34:55 GMT" } ]
2023-03-30T00:00:00
[ [ "Shen", "Li-Hsiang", "" ], [ "Su", "Chia-Yu", "" ], [ "Feng", "Kai-Ten", "" ] ]
TITLE: CoMP Enhanced Subcarrier and Power Allocation for Multi-Numerology based 5G-NR Networks ABSTRACT: With proliferation of fifth generation (5G) new radio (NR) technology, it is expected to meet the requirement of diverse traffic demands. We have designed a coordinated multi-point (CoMP) enhanced flexible multi-numerology (MN) for 5G-NR networks to improve the network performance in terms of throughput and latency. We have proposed a CoMP enhanced joint subcarrier and power allocation (CESP) scheme which aims at maximizing sum rate under the considerations of transmit power limitation and guaranteed quality-of-service (QoS) including throughput and latency restrictions. By employing difference of two concave functions (D.C.) approximation and abstract Lagrangian duality method, we theoretically transform the original non-convex nonlinear problem into a solvable maximization problem. Moreover, the convergence of our proposed CESP algorithm with D.C. approximation is analytically derived with proofs, and is further validated via numerical results. Simulation results demonstrated that our proposed CESP algorithm outperforms the conventional non-CoMP and single numerology mechanisms along with other existing benchmarks in terms of lower latency and higher throughput under the scenarios of uniform and edge users.
no_new_dataset
0.709856
2203.00219
Muhammad Akbar Husnoo
Muhammad Akbar Husnoo, Adnan Anwar, Nasser Hosseinzadeh, Shama Naz Islam, Abdun Naser Mahmood, Robin Doss
FedREP: Towards Horizontal Federated Load Forecasting for Retail Energy Providers
Appeared at 14th IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC 2022)
null
10.1109/APPEEC53445.2022.10072290
null
cs.DC cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
As Smart Meters are collecting and transmitting household energy consumption data to Retail Energy Providers (REP), the main challenge is to ensure the effective use of fine-grained consumer data while ensuring data privacy. In this manuscript, we tackle this challenge for energy load consumption forecasting in regards to REPs which is essential to energy demand management, load switching and infrastructure development. Specifically, we note that existing energy load forecasting is centralized, which are not scalable and most importantly, vulnerable to data privacy threats. Besides, REPs are individual market participants and liable to ensure the privacy of their own customers. To address this issue, we propose a novel horizontal privacy-preserving federated learning framework for REPs energy load forecasting, namely FedREP. We consider a federated learning system consisting of a control centre and multiple retailers by enabling multiple REPs to build a common, robust machine learning model without sharing data, thus addressing critical issues such as data privacy, data security and scalability. For forecasting, we use a state-of-the-art Long Short-Term Memory (LSTM) neural network due to its ability to learn long term sequences of observations and promises of higher accuracy with time-series data while solving the vanishing gradient problem. Finally, we conduct extensive data-driven experiments using a real energy consumption dataset. Experimental results demonstrate that our proposed federated learning framework can achieve sufficient performance in terms of MSE ranging between 0.3 to 0.4 and is relatively similar to that of a centralized approach while preserving privacy and improving scalability.
[ { "version": "v1", "created": "Tue, 1 Mar 2022 04:16:19 GMT" }, { "version": "v2", "created": "Tue, 28 Mar 2023 12:58:23 GMT" } ]
2023-03-30T00:00:00
[ [ "Husnoo", "Muhammad Akbar", "" ], [ "Anwar", "Adnan", "" ], [ "Hosseinzadeh", "Nasser", "" ], [ "Islam", "Shama Naz", "" ], [ "Mahmood", "Abdun Naser", "" ], [ "Doss", "Robin", "" ] ]
TITLE: FedREP: Towards Horizontal Federated Load Forecasting for Retail Energy Providers ABSTRACT: As Smart Meters are collecting and transmitting household energy consumption data to Retail Energy Providers (REP), the main challenge is to ensure the effective use of fine-grained consumer data while ensuring data privacy. In this manuscript, we tackle this challenge for energy load consumption forecasting in regards to REPs which is essential to energy demand management, load switching and infrastructure development. Specifically, we note that existing energy load forecasting is centralized, which are not scalable and most importantly, vulnerable to data privacy threats. Besides, REPs are individual market participants and liable to ensure the privacy of their own customers. To address this issue, we propose a novel horizontal privacy-preserving federated learning framework for REPs energy load forecasting, namely FedREP. We consider a federated learning system consisting of a control centre and multiple retailers by enabling multiple REPs to build a common, robust machine learning model without sharing data, thus addressing critical issues such as data privacy, data security and scalability. For forecasting, we use a state-of-the-art Long Short-Term Memory (LSTM) neural network due to its ability to learn long term sequences of observations and promises of higher accuracy with time-series data while solving the vanishing gradient problem. Finally, we conduct extensive data-driven experiments using a real energy consumption dataset. Experimental results demonstrate that our proposed federated learning framework can achieve sufficient performance in terms of MSE ranging between 0.3 to 0.4 and is relatively similar to that of a centralized approach while preserving privacy and improving scalability.
no_new_dataset
0.709007
2203.02194
Yibo Zhou
Yibo Zhou
Rethinking Reconstruction Autoencoder-Based Out-of-Distribution Detection
Accepted('Poster' presentation) as main conference paper of CVPR2022
in 2022 IEEE Conference on Computer Vision and Pattern Recognition
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In some scenarios, classifier requires detecting out-of-distribution samples far from its training data. With desirable characteristics, reconstruction autoencoder-based methods deal with this problem by using input reconstruction error as a metric of novelty vs. normality. We formulate the essence of such approach as a quadruplet domain translation with an intrinsic bias to only query for a proxy of conditional data uncertainty. Accordingly, an improvement direction is formalized as maximumly compressing the autoencoder's latent space while ensuring its reconstructive power for acting as a described domain translator. From it, strategies are introduced including semantic reconstruction, data certainty decomposition and normalized L2 distance to substantially improve original methods, which together establish state-of-the-art performance on various benchmarks, e.g., the FPR@95%TPR of CIFAR-100 vs. TinyImagenet-crop on Wide-ResNet is 0.2%. Importantly, our method works without any additional data, hard-to-implement structure, time-consuming pipeline, and even harming the classification accuracy of known classes.
[ { "version": "v1", "created": "Fri, 4 Mar 2022 09:04:55 GMT" }, { "version": "v2", "created": "Wed, 13 Apr 2022 10:02:58 GMT" }, { "version": "v3", "created": "Mon, 27 Mar 2023 03:12:05 GMT" }, { "version": "v4", "created": "Tue, 28 Mar 2023 01:12:24 GMT" }, { "version": "v5", "created": "Wed, 29 Mar 2023 02:13:23 GMT" } ]
2023-03-30T00:00:00
[ [ "Zhou", "Yibo", "" ] ]
TITLE: Rethinking Reconstruction Autoencoder-Based Out-of-Distribution Detection ABSTRACT: In some scenarios, classifier requires detecting out-of-distribution samples far from its training data. With desirable characteristics, reconstruction autoencoder-based methods deal with this problem by using input reconstruction error as a metric of novelty vs. normality. We formulate the essence of such approach as a quadruplet domain translation with an intrinsic bias to only query for a proxy of conditional data uncertainty. Accordingly, an improvement direction is formalized as maximumly compressing the autoencoder's latent space while ensuring its reconstructive power for acting as a described domain translator. From it, strategies are introduced including semantic reconstruction, data certainty decomposition and normalized L2 distance to substantially improve original methods, which together establish state-of-the-art performance on various benchmarks, e.g., the FPR@95%TPR of CIFAR-100 vs. TinyImagenet-crop on Wide-ResNet is 0.2%. Importantly, our method works without any additional data, hard-to-implement structure, time-consuming pipeline, and even harming the classification accuracy of known classes.
no_new_dataset
0.708792
2205.01265
Adish Singla
Adish Singla, Nikitas Theodoropoulos
From {Solution Synthesis} to {Student Attempt Synthesis} for Block-Based Visual Programming Tasks
Longer version of EDM 2022 paper
null
null
null
cs.AI cs.CY cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Block-based visual programming environments are increasingly used to introduce computing concepts to beginners. Given that programming tasks are open-ended and conceptual, novice students often struggle when learning in these environments. AI-driven programming tutors hold great promise in automatically assisting struggling students, and need several components to realize this potential. We investigate the crucial component of student modeling, in particular, the ability to automatically infer students' misconceptions for predicting (synthesizing) their behavior. We introduce a novel benchmark, StudentSyn, centered around the following challenge: For a given student, synthesize the student's attempt on a new target task after observing the student's attempt on a fixed reference task. This challenge is akin to that of program synthesis; however, instead of synthesizing a {solution} (i.e., program an expert would write), the goal here is to synthesize a {student attempt} (i.e., program that a given student would write). We first show that human experts (TutorSS) can achieve high performance on the benchmark, whereas simple baselines perform poorly. Then, we develop two neuro/symbolic techniques (NeurSS and SymSS) in a quest to close this gap with TutorSS.
[ { "version": "v1", "created": "Tue, 3 May 2022 01:32:47 GMT" }, { "version": "v2", "created": "Mon, 20 Jun 2022 21:42:16 GMT" }, { "version": "v3", "created": "Wed, 29 Mar 2023 00:07:18 GMT" } ]
2023-03-30T00:00:00
[ [ "Singla", "Adish", "" ], [ "Theodoropoulos", "Nikitas", "" ] ]
TITLE: From {Solution Synthesis} to {Student Attempt Synthesis} for Block-Based Visual Programming Tasks ABSTRACT: Block-based visual programming environments are increasingly used to introduce computing concepts to beginners. Given that programming tasks are open-ended and conceptual, novice students often struggle when learning in these environments. AI-driven programming tutors hold great promise in automatically assisting struggling students, and need several components to realize this potential. We investigate the crucial component of student modeling, in particular, the ability to automatically infer students' misconceptions for predicting (synthesizing) their behavior. We introduce a novel benchmark, StudentSyn, centered around the following challenge: For a given student, synthesize the student's attempt on a new target task after observing the student's attempt on a fixed reference task. This challenge is akin to that of program synthesis; however, instead of synthesizing a {solution} (i.e., program an expert would write), the goal here is to synthesize a {student attempt} (i.e., program that a given student would write). We first show that human experts (TutorSS) can achieve high performance on the benchmark, whereas simple baselines perform poorly. Then, we develop two neuro/symbolic techniques (NeurSS and SymSS) in a quest to close this gap with TutorSS.
no_new_dataset
0.652684
2206.06424
Mo Alloulah
Mohammed Alloulah, Maximilian Arnold
Look, Radiate, and Learn: Self-Supervised Localisation via Radio-Visual Correspondence
To appear in IEEE/CVF CVPR '23
null
null
null
cs.LG cs.NI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Next generation cellular networks will implement radio sensing functions alongside customary communications, thereby enabling unprecedented worldwide sensing coverage outdoors. Deep learning has revolutionised computer vision but has had limited application to radio perception tasks, in part due to lack of systematic datasets and benchmarks dedicated to the study of the performance and promise of radio sensing. To address this gap, we present MaxRay: a synthetic radio-visual dataset and benchmark that facilitate precise target localisation in radio. We further propose to learn to localise targets in radio without supervision by extracting self-coordinates from radio-visual correspondence. We use such self-supervised coordinates to train a radio localiser network. We characterise our performance against a number of state-of-the-art baselines. Our results indicate that accurate radio target localisation can be automatically learned from paired radio-visual data without labels, which is important for empirical data. This opens the door for vast data scalability and may prove key to realising the promise of robust radio sensing atop a unified communication-perception cellular infrastructure. Dataset will be hosted on IEEE DataPort.
[ { "version": "v1", "created": "Mon, 13 Jun 2022 19:08:36 GMT" }, { "version": "v2", "created": "Tue, 4 Oct 2022 21:54:35 GMT" }, { "version": "v3", "created": "Sun, 13 Nov 2022 21:34:39 GMT" }, { "version": "v4", "created": "Wed, 29 Mar 2023 10:11:26 GMT" } ]
2023-03-30T00:00:00
[ [ "Alloulah", "Mohammed", "" ], [ "Arnold", "Maximilian", "" ] ]
TITLE: Look, Radiate, and Learn: Self-Supervised Localisation via Radio-Visual Correspondence ABSTRACT: Next generation cellular networks will implement radio sensing functions alongside customary communications, thereby enabling unprecedented worldwide sensing coverage outdoors. Deep learning has revolutionised computer vision but has had limited application to radio perception tasks, in part due to lack of systematic datasets and benchmarks dedicated to the study of the performance and promise of radio sensing. To address this gap, we present MaxRay: a synthetic radio-visual dataset and benchmark that facilitate precise target localisation in radio. We further propose to learn to localise targets in radio without supervision by extracting self-coordinates from radio-visual correspondence. We use such self-supervised coordinates to train a radio localiser network. We characterise our performance against a number of state-of-the-art baselines. Our results indicate that accurate radio target localisation can be automatically learned from paired radio-visual data without labels, which is important for empirical data. This opens the door for vast data scalability and may prove key to realising the promise of robust radio sensing atop a unified communication-perception cellular infrastructure. Dataset will be hosted on IEEE DataPort.
new_dataset
0.715399
2206.11110
Aravinda Ramakrishnan Srinivasan
Aravinda Ramakrishnan Srinivasan, Yi-Shin Lin, Morris Antonello, Anthony Knittel, Mohamed Hasan, Majd Hawasly, John Redford, Subramanian Ramamoorthy, Matteo Leonetti, Jac Billington, Richard Romano, Gustav Markkula
Beyond RMSE: Do machine-learned models of road user interaction produce human-like behavior?
This work has been accepted for publication in the IEEE Transactions on Intelligent Transportation Systems journal on 13th March 2023
null
10.1109/TITS.2023.3263358
null
cs.LG cs.HC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Autonomous vehicles use a variety of sensors and machine-learned models to predict the behavior of surrounding road users. Most of the machine-learned models in the literature focus on quantitative error metrics like the root mean square error (RMSE) to learn and report their models' capabilities. This focus on quantitative error metrics tends to ignore the more important behavioral aspect of the models, raising the question of whether these models really predict human-like behavior. Thus, we propose to analyze the output of machine-learned models much like we would analyze human data in conventional behavioral research. We introduce quantitative metrics to demonstrate presence of three different behavioral phenomena in a naturalistic highway driving dataset: 1) The kinematics-dependence of who passes a merging point first 2) Lane change by an on-highway vehicle to accommodate an on-ramp vehicle 3) Lane changes by vehicles on the highway to avoid lead vehicle conflicts. Then, we analyze the behavior of three machine-learned models using the same metrics. Even though the models' RMSE value differed, all the models captured the kinematic-dependent merging behavior but struggled at varying degrees to capture the more nuanced courtesy lane change and highway lane change behavior. Additionally, the collision aversion analysis during lane changes showed that the models struggled to capture the physical aspect of human driving: leaving adequate gap between the vehicles. Thus, our analysis highlighted the inadequacy of simple quantitative metrics and the need to take a broader behavioral perspective when analyzing machine-learned models of human driving predictions.
[ { "version": "v1", "created": "Wed, 22 Jun 2022 14:04:39 GMT" }, { "version": "v2", "created": "Tue, 28 Mar 2023 18:38:59 GMT" } ]
2023-03-30T00:00:00
[ [ "Srinivasan", "Aravinda Ramakrishnan", "" ], [ "Lin", "Yi-Shin", "" ], [ "Antonello", "Morris", "" ], [ "Knittel", "Anthony", "" ], [ "Hasan", "Mohamed", "" ], [ "Hawasly", "Majd", "" ], [ "Redford", "John", "" ], [ "Ramamoorthy", "Subramanian", "" ], [ "Leonetti", "Matteo", "" ], [ "Billington", "Jac", "" ], [ "Romano", "Richard", "" ], [ "Markkula", "Gustav", "" ] ]
TITLE: Beyond RMSE: Do machine-learned models of road user interaction produce human-like behavior? ABSTRACT: Autonomous vehicles use a variety of sensors and machine-learned models to predict the behavior of surrounding road users. Most of the machine-learned models in the literature focus on quantitative error metrics like the root mean square error (RMSE) to learn and report their models' capabilities. This focus on quantitative error metrics tends to ignore the more important behavioral aspect of the models, raising the question of whether these models really predict human-like behavior. Thus, we propose to analyze the output of machine-learned models much like we would analyze human data in conventional behavioral research. We introduce quantitative metrics to demonstrate presence of three different behavioral phenomena in a naturalistic highway driving dataset: 1) The kinematics-dependence of who passes a merging point first 2) Lane change by an on-highway vehicle to accommodate an on-ramp vehicle 3) Lane changes by vehicles on the highway to avoid lead vehicle conflicts. Then, we analyze the behavior of three machine-learned models using the same metrics. Even though the models' RMSE value differed, all the models captured the kinematic-dependent merging behavior but struggled at varying degrees to capture the more nuanced courtesy lane change and highway lane change behavior. Additionally, the collision aversion analysis during lane changes showed that the models struggled to capture the physical aspect of human driving: leaving adequate gap between the vehicles. Thus, our analysis highlighted the inadequacy of simple quantitative metrics and the need to take a broader behavioral perspective when analyzing machine-learned models of human driving predictions.
no_new_dataset
0.709447