检索结果-内蒙古大学图书馆

FedSSP: Federated Graph Learning with Spectral Knowledge and Personalized Preference

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Tan, Zihan Wan, Guancheng Huang, Wenke Ye, Mang National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering School of Computer Science Wuhan University Wuhan China Taikang Center for Life and Medical Sciences Wuhan University Wuhan China

Personalized Federated Graph Learning (pFGL) facilitates the decentralized training of Graph Neural networks (GNNs) without compromising privacy while accommodating personalized requirements for non-IID participants. In cross-domain scenarios, structural heterogeneity poses significant challenges for pFGL. Nevertheless, previous pFGL methods incorrectly share non-generic knowledge globally and fail to tailor personalized solutions locally under domain structural shift. We innovatively reveal that the spectral nature of graphs can well reflect inherent domain structural shifts. Correspondingly, our method overcomes it by sharing generic spectral knowledge. Moreover, we indicate the biased message-passing schemes for graph structures and propose the personalized preference module. Combining both strategies, we propose our pFGL framework FedSSP which Shares generic Spectral knowledge while satisfying graph Preferences. Furthermore, We perform extensive experiments on cross-dataset and cross-domain settings to demonstrate the superiority of our framework. The code is available at https://***/OakleyTan/FedSSP. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

Good is Bad: Causality Inspired Cloth-debiasing for Cloth-changing Person Re-identification

学校读者我要写书评

暂无评论

Good is Bad: Causality Inspired Cloth-debiasing for Cloth-ch...

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Zhengwei Yang Meng Lin Xian Zhong Yu Wu Zheng Wang National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence School of Computer Science Wuhan University Hubei Key Laboratory of Multimedia and Network Communication Engineering School of Computer Science and Artificial Intelligence Wuhan University of Technology

Entangled representation of clothing and identity (ID)-intrinsic clues are potentially concomitant in conventional person Re- IDentification (ReID). Nevertheless, eliminating the negative impact of clothing on ID remains challenging due to the lack of theory and the difficulty of isolating the exact implications. In this paper, a causality-based Auto-Intervention Model, referred to as AIM 1 1 Codes will publicly available at https://***/BoomShakaY/AIM-CCReID., is first proposed to mitigate clothing bias for robust cloth-changing person ReID (CC-ReID). Specifically, we analyze the effect of clothing on the model inference and adopt a dual-branch model to simulate causal intervention. Progressively, clothing bias is eliminated automatically with model training. AIM is encouraged to learn more discriminative ID clues that are free from clothing bias. Extensive experiments on two standard CC-ReID datasets demonstrate the superiority of the proposed AIM over other state-of-the-art methods.

关键词：

Unlabeled Imperfect Demonstrations in Adversarial Imitation Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Yunke Du, Bo Xu, Chang School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan China School of Computer Science Faculty of Engineering The University of Sydney Australia

Adversarial imitation learning has become a widely used imitation learning framework. The discriminator is often trained by taking expert demonstrations and policy trajectories as examples respectively from two categories (positive vs. negative) and the policy is then expected to produce trajectories that are indistinguishable from the expert demonstrations. But in the real world, the collected expert demonstrations are more likely to be imperfect, where only an unknown fraction of the demonstrations are optimal. Instead of treating imperfect expert demonstrations as absolutely positive or negative, we investigate unlabeled imperfect expert demonstrations as they are. A positive-unlabeled adversarial imitation learning algorithm is developed to dynamically sample expert demonstrations that can well match the trajectories from the constantly optimized agent policy. The trajectories of an initial agent policy could be closer to those non-optimal expert demonstrations, but within the framework of adversarial imitation learning, agent policy will be optimized to cheat the discriminator and produce trajectories that are similar to those optimal expert demonstrations. Theoretical analysis shows that our method learns from the imperfect demonstrations via a self-paced way. Experimental results on MuJoCo and RoboSuite platforms demonstrate the effectiveness of our method from different aspects. Copyright © 2023, The Authors. All rights reserved.

关键词： Demonstrations

Better Diffusion Models Further Improve Adversarial Training

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Zekai Pang, Tianyu Du, Chao Lin, Min Liu, Weiwei Yan, Shuicheng School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China Sea AI Lab

It has been recognized that the data generated by the denoising diffusion probabilistic model (DDPM) improves adversarial training. After two years of rapid development in diffusion models, a question naturally arises: can better diffusion models further improve adversarial training? This paper gives an affirmative answer by employing the most recent diffusion model (Karras et al., 2022) which has higher efficiency (∼ 20 sampling steps) and image quality (lower FID score) compared with DDPM. Our adversarially trained models achieve state-of-the-art performance on RobustBench using only generated data (no external datasets). Under the ∞norm threat model with ϵ = 8/255, our models achieve 70.69% and 42.67% robust accuracy on CIFAR-10 and CIFAR-100, respectively, i.e. improving upon previous state-of-the-art models by +4.58% and +8.03%. Under the 2-norm threat model with ϵ = 128/255, our models achieve 84.86% on CIFAR-10 (+4.44%). These results also beat previous works that use external data. We also provide compelling results on the SVHN and TinyImageNet datasets. Our code is at https://***/wzekai99/DM-Improves-AT. Copyright © 2023, The Authors. All rights reserved.

关键词： Diffusion

Anomize: Better Open Vocabulary Video Anomaly Detection

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Li, Fei Liu, Wenxuan Chen, Jingjing Zhang, Ruixu Wang, Yuran Zhong, Xian Wang, Zheng National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University China State Key Laboratory for Multimedia Information Processing School of Computer Science Peking University China Hubei Key Laboratory of Transportation Internet of Things Wuhan University of Technology China

Open Vocabulary Video Anomaly Detection (OVVAD) seeks to detect and classify both base and novel anomalies. However, existing methods face two specific challenges related to novel anomalies. The first challenge is detection ambiguity, where the model struggles to assign accurate anomaly scores to unfamiliar anomalies. The second challenge is categorization confusion, where novel anomalies are often misclassified as visually similar base instances. To address these challenges, we explore supplementary information from multiple sources to mitigate detection ambiguity by leveraging multiple levels of visual data alongside matching textual information. Furthermore, we propose incorporating label relations to guide the encoding of new labels, thereby improving alignment between novel videos and their corresponding labels, which helps reduce categorization confusion. The resulting Anomize framework effectively tackles these issues, achieving superior performance on UCF-CRIME and XD-VIOLENCE datasets, demonstrating its effectiveness in OVVAD. Copyright © 2025, The Authors. All rights reserved.

关键词：

Robust Synthetic-to-Real Transfer for Stereo Matching

学校读者我要写书评

暂无评论

Robust Synthetic-to-Real Transfer for Stereo Matching

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Jiawei Zhang Jiahe Li Lei Huang Xiaohan Yu Lin Gu Jin Zheng Xiao Bai State Key Laboratory of Complex & Critical Software Environment School of Computer Science and Engineering Jiangxi Research Institute Beihang University SKLCCSE Institute of Artificial Intelligence Beihang University School of Computing Macquarie University Australia RIKEN AIP The University of Tokyo

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

With advancements in domain generalized stereo matching networks, models pre-trained on synthetic data demonstrate strong robustness to unseen domains. However, few studies have investigated the robustness after fine-tuning them in real-world scenarios, during which the domain generalization ability can be seriously degraded. In this paper, we explore fine-tuning stereo matching networks without compromising their robustness to unseen domains. Our motivation stems from comparing Ground Truth (GT) versus Pseudo Label (PL) for fine-tuning: GT degrades, but PL preserves the domain generalization ability. Empirically, we find the difference between GT and PL implies valuable information that can regularize networks during fine-tuning. We also propose a framework to utilize this difference for fine-tuning, consisting of a frozen Teacher, an exponential moving average (EMA) Teacher, and a Student network. The core idea is to utilize the EMA Teacher to measure what the Student has learned and dynamically improve GT and PL for fine-tuning. We integrate our framework with state-of-the-art networks and evaluate its effectiveness on several real-world datasets. Extensive experiments show that our method effectively preserves the domain generalization ability during fine-tuning. Code is available at: https://***/jiaw-z/DKT-Stereo.

关键词： computer vision Codes Robustness Data models Pattern matching Synthetic data Meteorology

DRSC: Dual-Reweighted Siamese Contrastive Learning network for Cross-Domain Rotating Machinery Fault Diagnosis With Multi-Source Domain Imbalanced Data

学校读者我要写书评

暂无评论

IEEE Internet of Things Journal 2025年

作者： Bi, Yuanguo Fu, Rao Jiang, Cunyu Zhang, Xiaoling Li, Fengyun Zhao, Liang Han, Guangjie Northeastern University School of Computer Science and Engineering Shenyang110169 China Ministry of Education Engineering Research Center of Security Technology of Complex Network System Shenyang110169 China Shenyang University of Technology Institute of Artificial Intelligence Shenyang110870 China Shenyang Aerospace University School of Computer Science Shenyang110135 China Hohai University Key Laboratory of Maritime Intelligent Network Information Technology Ministry of Education China

To enhance the reliability of rotating machinery, cross-domain fault diagnosis becomes vital for detecting faults under unknown operating conditions. However, multi-source domain imbalanced data present significant challenges, as divergent label distributions across domains cause complex domain-class shifts and degrade the performance of cross-domain fault diagnosis. Moreover, diagnostic models often struggle to learn features from minority classes due to label imbalance within each domain, which may degrade the performance in diagnosing these minority classes. To address these challenges, we propose a Dual-Reweighted Siamese Contrastive learning network (DRSC) for cross-domain fault diagnosis with multi-source domain imbalanced data. In DRSC, we design a Siamese feature extractor based on a wide-kernel convolutional neural network to capture short-term characteristics and leverage the convenience in extracting domain-invariant features. Subsequently, to alleviate domain-class shifts, we design a reweighted contrastive domain-class alignment mechanism that strategically pulls domain-class pairs together while pushing other health conditions away. Finally, to enable the diagnostic model to learn from minority health conditions, a reweighted health condition classifier is developed by assigning higher weights to the minority classes. Evaluation results on two public datasets illustrate DRSC outperforms comparison models in cross-domain fault diagnosis. © 2014 IEEE.

关键词： Convolutional neural networks

SAMRS: scaling-up remote sensing segmentation dataset with segment anything model 23

学校读者我要写书评

暂无评论

SAMRS: scaling-up remote sensing segmentation dataset with s...

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Di Wang Jing Zhang Bo Du Minqiang Xu Lin Liu Dacheng Tao Liangpei Zhang School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence and Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China School of Computer Science Faculty of Engineering The University of Sydney Australia National Engineering Research Center of Speech and Language Information Processing China State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing Wuhan University China

The success of the Segment Anything Model (SAM) demonstrates the significance of data-centric machine learning. However, due to the difficulties and high costs associated with annotating Remote Sensing (RS) images, a large amount of valuable RS data remains unlabeled, particularly at the pixel level. In this study, we leverage SAM and existing RS object detection datasets to develop an efficient pipeline for generating a large-scale RS segmentation dataset, dubbed SAMRS. SAMRS totally possesses 105,090 images and 1,668,241 instances, surpassing existing high-resolution RS segmentation datasets in size by several orders of magnitude. It provides object category, location, and instance information that can be used for semantic segmentation, instance segmentation, and object detection, either individually or in combination. We also provide a comprehensive analysis of SAMRS from various aspects. Moreover, preliminary experiments highlight the importance of conducting segmentation pre-training with SAMRS to address task discrepancies and alleviate the limitations posed by limited training data during fine-tuning. The code and dataset will be available at https://***/ViTAE-Transformer/SAMRS.

关键词：

SAMRS: Scaling-up Remote Sensing Segmentation Dataset with Segment Anything Model

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Di Zhang, Jing Du, Bo Xu, Minqiang Liu, Lin Tao, Dacheng Zhang, Liangpei School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China School of Computer Science Faculty of Engineering The University of Sydney Australia National Engineering Research Center of Speech and Language Information Processing China State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing Wuhan University China

关键词： Remote sensing