检索结果-内蒙古大学图书馆

16th International Conference on Cyber-Enabled Distributed Computing and Knowledge Discover, CyberC 2024

作者： Chen, Yan Zhao, Sunhang Wang, Xinyuan Jin, Zhengxuan Lin, Zekai Wang, Bin The Third Research Institute of the Ministry of Public Security Shanghai China College of Letters and Science United States Zhejiang Gongshang University Computer Science and Technology China Binjiang Institute of Zhejiang University China Zhengzhou University School of Computer and Artificial Intelligence China Zhejiang Key Laboratory of Multi-Dimensional Perception Technology Application and Cybersecurity China

ISBN: (纸本)9798331506896

This paper presents RAS-Rec, a novel and robust approach to sequence recommendation that leverages the power of data augmentation techniques. RAS-Rec proposes data augmentation as a means to enhance the representations of users and items in sequence recommendation. By generating synthetic training samples through various transformations applied to the original data, RAS-Rec increases the diversity and quantity of the training set. The augmented sequence data addresses the limitations of existing methods and improves the robustness and generalization capabilities of sequence recommendation models. Additionally, a contrastive learning scheme is designed to train and further enhance the representations in RAS-Rec. Extensive experiments validate the effectiveness of the proposed approach, demonstrating its superiority over existing methods in terms of recommendation performance. The results highlight the value of data augmentation and contrastive learning in sequence recommendation tasks. The findings of this study contribute to the advancement of robust sequence recommendation techniques. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Privacy-Preserving Large-Scale AI Models for Intelligent Railway Transportation Systems:Hierarchical Poisoning Attacks and Defenses in Federated Learning

引用

Computer Modeling in Engineering & Sciences 2024年第11期141卷 1305-1325页

作者： Yongsheng Zhu Chong Liu Chunlei Chen Xiaoting Lyu Zheng Chen Bin Wang Fuqiang Hu Hanxi Li Jiao Dai Baigen Cai Wei Wang School of Automation and Intelligence Beijing Jiaotong UniversityBeijing100044China Institute of Computing Technologies China Academy of Railway Sciences Corporation LimitedBeijing100081China School of Computer Science and Technology Beijing Jiaotong UniversityBeijing100044China Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing Jiaotong UniversityBeijing100044China Institute of Infrastructure Inspection China Academy of Railway Sciences Corporation LimitedBeijing100081China Zhejiang Key Laboratory of Multi-Dimensional Perception Technology Application and CybersecurityHangzhou310053China

The development of Intelligent Railway Transportation Systems necessitates incorporating privacy-preserving mechanisms into AI models to protect sensitive information and enhance system *** learning offers a promising solution by allowing multiple clients to train models collaboratively without sharing private ***,despite its privacy benefits,federated learning systems are vulnerable to poisoning attacks,where adversaries alter local model parameters on compromised clients and send malicious updates to the server,potentially compromising the global model’s *** this study,we introduce PMM(Perturbation coefficient multiplied by Maximum value),a new poisoning attack method that perturbs model updates layer by layer,demonstrating the threat of poisoning attacks faced by federated *** experiments across three distinct datasets have demonstrated PMM’s ability to significantly reduce the global model’s ***,we propose an effective defense method,namely CLBL(Cluster Layer By Layer).Experiment results on three datasets have confirmed CLBL’s effectiveness.

关键词： Privacy-preserving intelligent railway transportation system federated learning poisoning attacks defenses

来源：评论

学校读者我要写书评

暂无评论

Hey! You Photographed My Screen without Approved: A Practical Screen Content Protection Scheme Based on Visual Cryptography

Hey! You Photographed My Screen without Approved: A Practica...

引用

2022 International Conference on Networking and Network applications, NaNA 2022

作者： Liu, Kui Wang, Bin Shi, Ziwei Chen, Jiadong Zhang, Zhiwei School of Computer Science and Technology Xidian University Shaanxi Xi’an710071 China Network and Information Security Laboratory Hangzhou Hikvision Digital Technology Co. LTD. Hangzhou China Zhejiang Key Laboratory of Multi-dimensional Perception Technology Application and Cybersecurity Hangzhou310053 China

ISBN: (数字)9781665461313

ISBN: (纸本)9781665461313

Nowadays , with the continuous development of digital media, electronic screens have become the most direct tool for people to interact with information. Various styles of electronic screens provide vivid information exchange methods for people's lives, and at the same time make information leakage more likely to happen. Electronic screens are susceptible to stealing photographs when sharing information, which leads to information leakage and security issues. Under the unavoidable situation of such leaks, various protection schemes have been proposed before, but these schemes cannot effectively protect the screen content. Therefore, in this paper, we propose an effective a screen content protection scheme, based on Visual Cryptography(VC) for image encryption, combined with Scale-Invariant Feature Transform(SIFT) to deal with screen stealing scenes. In addition, our experiments show that our scheme is superior to existing schemes in terms of convenience and security. © 2022 IEEE.

关键词： Digital storage

来源：评论

学校读者我要写书评

暂无评论

PAGE: Equilibrate Personalization and Generalization in Federated Learning

arXiv

引用

arXiv 2023年

作者： Chen, Qian Wang, Zilong Hu, Jiaqi Yan, Haonan Zhou, Jianying Lin, Xiaodong State Key Laboratory of Integrated Service Networks Xidian University Xi’an China Zhejiang Key Laboratory of Multi-dimensional Perception Technology Application and Cybersecurity Zhejiang China iTrust Singapore University of Technology and Design Singapore Singapore School of Computer Science University of Guelph Guelph Canada

Federated learning (FL) is becoming a major driving force behind machine learning as a service, where customers (clients) collaboratively benefit from shared local updates under the orchestration of the service provider (server). Representing clients’ current demands and the server’s future demand, local model personalization and global model generalization are separately investigated, as the ill-effects of data heterogeneity enforce the community to focus on one over the other. However, these two seemingly competing goals are of equal importance rather than black and white issues, and should be achieved simultaneously. In this paper, we propose the first algorithm to balance personalization and generalization on top of game theory, dubbed PAGE, which reshapes FL as a co-opetition game between clients and the server. To explore the equilibrium, PAGE further formulates the game as Markov decision processes, and leverages the reinforcement learning algorithm, which simplifies the solving complexity. Extensive experiments on four widespread datasets show that PAGE outperforms state-of-the-art FL baselines in terms of global and local prediction accuracy simultaneously, and the accuracy can be improved by up to 35.20% and 39.91%, respectively. In addition, biased variants of PAGE imply promising adaptiveness to demand shifts in practice. Copyright © 2023, The Authors. All rights reserved.

关键词： Game theory

来源：评论

学校读者我要写书评

暂无评论

Towards robust data augmentation for sequence recommendation

Towards robust data augmentation for sequence recommendation

引用

International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC

作者： Yan Chen Sunhang Zhao Xinyuan Wang Zhengxuan Jin Zekai Lin Bin Wang The Third Research Institute of the Ministry of Public Security Shanghai China College of Letters and Science University of California Santa Barbara (UCSB) Computer Science and Technology Zhejiang Gongshang University China Binjiang Institute of Zhejiang University China School of Computer and Artificial Intelligence Zhengzhou University China Zhejiang Key Laboratory of Multi-Dimensional Perception Technology Application and Cybersecurity China

ISBN: (数字)9798331506896

ISBN: (纸本)9798331506902

关键词：

来源：评论

学校读者我要写书评

暂无评论

Two-in-One Solution: Simultaneously Enhancing Security and Privacy for Data-Driven Models in Mobile Edge Computing

引用

IEEE Transactions on Consumer Electronics 2024年

作者： Liu, Pengrui Yuan, Xiaohan Wang, Wei Xu, Xiangrui Li, Tao Wang, Junyong Wang, Bin Pedrycz, Witold Beijing Jiaotong University Beijing Key Laboratory of Security and Privacy in Intelligent Transportation 3 Shangyuancun Beijing100044 China Haihe Lab of ITAI Tianjin300450 China Zhejiang Key Laboratory of Multi-dimensional Perception Technology Application and Cybersecurity Hangzhou310053 China University of Alberta Department of Electrical and Computer Engineering EdmontonABT6G 2R3 Canada

Data-driven models are widely employed in Mobile Edge Computing to satisfy the demands of Emerging Consumer applications. However, previous work demonstrates that data-driven models are susceptible to security threats like Backdoor and Evasion Attacks or privacy threats like Membership Inference Attacks. Numerous existing methods for mitigating these threats have been proposed. However, these methods focus solely on enhancing model security or only on improving model privacy. Ideally, data-driven models should enhance model security and privacy simultaneously. In this paper, we propose methods that combine individual security-enhancing and privacy-enhancing methods to mitigate the security and privacy threats of data-driven models simultaneously. We evaluate the effectiveness of individual security-enhancing methods, individual privacy-enhancing methods, and our methods in simultaneously enhancing model security and privacy. Our comprehensive experimental analysis reveals two-fold insights. First, individual security-enhancing methods can either enhance or diminish model privacy, while individual privacy-enhancing methods face challenges in enhancing model security. Second, our methods improve the effectiveness of simultaneously enhancing model security and privacy compared to the individual methods. © 1975-2011 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Bfs2adv: Black-Box Adversarial Attack Towards Hard-to-Attack Short Texts

SSRN

引用

SSRN 2023年

作者： Han, Xu Li, Qiang Cao, Hongbo Han, Lei Wang, Bin Bao, Xuhua Han, Yufei Wang, Wei Beijing Key Laboratory of Security and Privacy of Intelligent Transportation Beijing Jiaotong University Beijing100044 China Beijing Institute of Computer Technology and Application Beijing100584 China Zhejiang Key Laboratory of Multi-dimensional Perception Technology Zhejiang University Hangzhou310027 China Sangfro Technologies Inc Shenzhen518055 China INRIA Bretagne Rennes35042 France

The advent of Machine Learning as a Service (MLaaS) and deep learning applications has increased the susceptiblility of models to adversarial textual attacks, particularly in black-box settings. Prior work on black-box adversarial textual attacks generally follows a stable strategy that involves leveraging char-level, world-level, and sentence-level perturbations, as well as using queries to the target model to find adversarial examples in the search space. However, existing approaches prioritize query efficiency by reducing the search space, thereby overlooking hard-to-attack textual instances. To address this issue, we propose \textit{BFS2Adv}, a brute force algorithm that generates adversarial examples for both easy-to-attack and hard-to-attack textual ***2Adv, starting with an original text, employs word-level perturbations and synonym substitution to construct a comprehensive search space, with each node representing a potential adversarial example. The algorithm systematically explores this space through a breadth-first search, combined with queries to the target model, to effectively identify qualified adversarial *** implemented and evaluated a prototype of BFS2Adv against renowned models such as ALBERT and BERT, utilizing the SNLI and MR datasetsOur results demonstrate that BFS2Adv outperforms state-of-the-art algorithms and effectively improves the success rate of short-text adversarial ***, we provide detailed insights into the robustness of BFS2Adv by analyzing those hard-to-attack examples. © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Hessian-Free Second-Order Adversarial Examples for Adversarial Learning

arXiv

引用

arXiv 2022年

作者： Qian, Yaguan Wang, Yuqi Wang, Bin Gu, Zhaoquan Guo, Yuhan Swaileh, Wassim School of Big Data Science Zhejiang University of Science and Technology Hangzhou China Zhejiang Key Laboratory of Multi-dimensional Perception Technology Application and Cybersecurity Hikvision Digital Technology Co. Ltd Hangzhou China Cyberspace Institute of Advanced Technology Guangzhou University Guangzhou China ETIS Research Laboratory CY Cergy Paris University Paris France

—Recent studies show deep neural networks (DNNs) are extremely vulnerable to the elaborately designed adversarial examples. Adversarial learning with those adversarial examples has been proved as one of the most effective methods to defend against such an attack. At present, most existing adversarial examples generation methods are based on first-order gradients, which can hardly further improve models’ robustness, especially when facing second-order adversarial attacks. Compared with first-order gradients, second-order gradients provide a more accurate approximation of the loss landscape with respect to natural examples. Inspired by this, our work crafts second-order adversarial examples and uses them to train DNNs. Nevertheless, second-order optimization involves time-consuming calculation for Hessian-inverse. We propose an approximation method through transforming the problem into an optimization in the Krylov subspace, which remarkably reduce the computational complexity to speed up the training procedure. Extensive experiments conducted on the MINIST and CIFAR-10 datasets show that our adversarial learning with second-order adversarial examples outperforms other fisrt-order methods, which can improve the model robustness against a wide range of attacks. Copyright © 2022, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

[Retracted] Defending Privacy Inference Attacks to Federated Learning for Intelligent IoT with Parameter Compression

引用

Security and Communication Networks 2023年第1期2023卷

作者： Yongsheng Zhu Hongbo Cao Yuange Ren Wanqi Wang Bin Wang Mingqing Hu Baigen Cai Wei Wang School of Electronic and Information Engineering Beijing Jiaotong University No. 3 Shangyuancun Beijing 100044 *** Institute of Computing Technologies China Academy of Railway Sciences Corporation Limited Beijing 100081 *** Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing Jiaotong University No. 3 Shangyuancun Beijing 100044 *** Zhejiang Key Laboratory of Multi-Dimensional Perception Technology Application and Cybersecurity Hangzhou 310053 China iFLYTEK Co. Ltd. Hefei China

Federated learning has been popularly studied with people’s increasing awareness of privacy protection. It solves the problem of privacy leakage by its ability that allows many clients to train a collaborative model without uploading local data collected by Internet of Things (IoT) devices. However, there are still threats of privacy leakage in federated learning. The privacy inference attacks can reconstruct the privacy data of other clients based on GAN from the parameters in the process of iterations for global models. In this work, we are motivated to prevent GAN-based privacy inference attacks in federated learning. Inspired by the idea of gradient compression, we propose a defense method called Federated Learning Parameter Compression (FLPC) which can reduce the sharing of information for privacy protection. It prevents attackers from recovering the private information of victims while maintaining the accuracy of the global model. Extensive experimental results demonstrated that our method is effective in the prevention of GAN-based privacy inferring attacks. In addition, based on the experimental results, we propose a norm-based metric to assess the performance of privacy-preserving.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：