检索结果-内蒙古大学图书馆

VFLMonitor: Defending One-Party Hijacking Attacks in Vertical Federated Learning

IEEE Transactions on Information Forensics and security 2025年 20卷 4828-4843页

作者： Xu, Xiangrui Zhao, Yiwen Han, Yufei Zhu, Yongsheng Han, Zhen Xu, Guangquan Wang, Bin Ji, Shouling Wang, Wei Beijing Jiaotong University Beijing Key Laboratory of Security and Privacy in Intelligent Transportation Beijing100044 China University of Southern California Thomas Lord Department of Computer Science Los Angeles90007 United States INRIA Bretagne Rennes35042 France College of Intelligence and Computing Tianjin300072 China Network and Data Security Hangzhou310053 China Zhejiang University College of Computer Science and Technology Zhejiang 310058 China Xi’an Jiaotong University Ministry of Education Key Lab for Intelligent Networks and Network Security Xian710049 China Beijing Jiaotong University school of computer science and technology Beijing100044 China

Vertical Federated Learning (VFL) is susceptible to various one-party hijacking attacks, such as Replay and Generation attacks, where a single malicious client can manipulate the model to produce attacker-specified results, thereby compromising its reliability in real-world deployments. In this paper, we first uncover the underlying mechanisms of these attacks and observe that successful attacks induce significant discrepancies in the embedding-label associations across different clients. We establish a theoretical framework demonstrating how these discrepancies can serve as reliable indicators for detecting hijacking attempts. Building upon this insight, we propose VFLMonitor, a robust defense mechanism that leverages these embedding-label discrepancies to detect and mitigate hijacking attacks. Specifically, VFLMonitor identifies suspicious queries by analyzing differences in label estimations from multiple clients and applies a majority voting rule to correct or filter out these malicious queries. Moreover, VFLMonitor introduces a novel regularization strategy during training to reduce intra-class variance in embeddings, thereby enhancing their discriminative power and improving defense effectiveness. Extensive experi21 ments were conducted on 5 real-world datasets against 2 different attack types under 3 attack scenarios. The results demonstrate that VFLMonitor can effectively identify and exclude potential hijacked requests in all types of one-party hijacking attacks, while maintaining a meager false positive rate for legitimate queries. © 2005-2012 IEEE.

关键词： Training Servers Electronic mail Artificial intelligence Reliability theory Predictive models Federated learning data models Computational modeling Collaboration

来源：评论

学校读者我要写书评

暂无评论

High-Performance Implementation Architecture for Additive Homomorphic Encryption Algorithms on FPGA

High-Performance Implementation Architecture for Additive Ho...

引用

IEEE International Conference on Big data security on Cloud (Bigdatasecurity), High Performance and Smart Computing (HPSC) and Intelligent data and security (IDS)

作者： Yuxuan Zhang Hua Guo Chen Junxin Chen Wenmao Liu School of Cyber Science and Technology Beihang University BeiJing China Key Laboratory of Data and Intelligent System Security Ministry of Education Tianjin China NSFOCUS Technologies Group Co. Ltd. BeiJing China

ISBN: (数字)9798331595104

ISBN: (纸本)9798331595111

Additive homomorphic encryption is one of the key technologies for ensure data privacy in federated learning. To accelerate these algorithms, several studies have deployed homomorphic encryption algorithms on parallel platforms, such as FPGAs. Although many works have demonstrated good acceleration effects, an efficient and generalizable implementation architecture remains lacking. In this paper, we propose an efficient FPGA implementation architecture compatible with various partially homomorphic encryption algorithms. First, an efficient modular multiplier which supports dynamic modulus length is designed. We introduce the redundant number system and design an iteration cycle control module to reduce computing cost and to fit the modulus with different length, respectively. Secondly, a high-performance FPGA implementation architecture compatible with multiple additive homomorphic encryption algorithms is proposed. Based on the processes of algorithms, we propose a customized multifunctional multiplier and a top-level control module to realize a full reuse without any performance loss. According to implementation results on Kintex UltraScale XCKU060, the encrypt speed of Paillier and OU achieves 473.79 OP/s and 88.96 OP/s under 128-bit security, respectively. Compared with existing similar works, our architecture achieves 8.4 × speedup, demonstrating its performance and functional advantages.

关键词： data privacy Additives Federated learning Heuristic algorithms Process control Computer architecture Hardware Cryptography Homomorphic encryption Field programmable gate arrays

来源：评论

学校读者我要写书评

暂无评论

NI-GDBA: Non-Intrusive Distributed Backdoor Attack Based on Adaptive Perturbation on Federated Graph Learning 34

NI-GDBA: Non-Intrusive Distributed Backdoor Attack Based on ...

引用

34th ACM Web Conference, WWW 2025

作者： Li, Ken Shi, Bin Wei, Jiazhe Dong, Bo School of Computer Science and Technology Xi’an Jiaotong University Ministry of Education Key Laboratory of Intelligent Networks and Network Security Xi’an China School of Computer Science and Technology Xi’an Jiaotong University Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi’an China School of Distance Education Xi’an Jiaotong University Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi’an China

ISBN: (纸本)9798400712746

Federated Graph Learning (FedGL) is an emerging Federated Learning (FL) framework that learns the graph data from various clients to train better Graph Neural networks(GNNs) model. Owing to concerns regarding the security of such framework, numerous studies have attempted to execute backdoor attacks on FedGL, with a particular focus on distributed backdoor attacks. However, all existing methods posting distributed backdoor attack on FedGL only focus on injecting distributed backdoor triggers into the training data of each malicious client, which will cause model performance degradation on original task and is not always effective when confronted with robust federated learning defense algorithms, leading to low success rate of attack. What’s more, the backdoor signals introduced by the malicious clients may be smoothed out by other clean signals from the honest clients, which potentially undermining the performance of the attack. To address the above significant shortcomings, we propose a non-intrusive graph distributed backdoor attack(NI-GDBA) that does not require backdoor triggers to be injected in the training data. Our attack trains an adaptive perturbation trigger generator model for each malicious client to learn the natural backdoor from the GNN model downloading from the server with the malicious client’s local data. In contrast to traditional distributed backdoor attacks on FedGL via trigger injection in training data, our attack on different datasets such as Molecules and Bioinformatics have higher attack success rate, stronger persistence and stealth, and has no negative impact on the performance of the global GNN model. We also explore the robustness of NI-GDBA under different defense strategies, and based on our extensive experimental studies, we show that our attack method is robust to current federated learning defense methods, thus it is necessary to consider non-intrusive distributed backdoor attacks on FedGL as a novel threat that requires custom d

关键词： Backdoor Attacks Federated Graph Learning

来源：评论

学校读者我要写书评

暂无评论

Towards an Ontology of Traceable Impact Management in the Food Supply Chain

arXiv

引用

arXiv 2025年

作者： Gajderowicz, Bart Fox, Mark S. Gao, Yongchao Urban Data Centre School of Cities University of Toronto Toronto Canada Key Laboratory of Computing Power Network and Information Security Shandong Computer Science Center Qilu University of Technology Shandong Academy of Sciences Jinan China

The pursuit of quality improvements and accountability in the food supply chains, especially how they relate to food-related outcomes, such as hunger, has become increasingly vital, necessitating a comprehensive approach that encompasses product quality and its impact on various stakeholders and their communities. Such an approach offers numerous benefits in increasing product quality and eliminating superfluous measurements while appraising and alleviating the broader societal and environmental repercussions. A traceable impact management model (TIMM) provides an impact structure and a reporting mechanism that identifies each stakeholder’s role in the total impact of food production and consumption stages. The model aims to increase traceability’s utility in understanding the impact of changes on communities affected by food production and consumption, aligning with current and future government requirements, and addressing the needs of communities and consumers. This holistic approach is further supported by an ontological model that forms the logical foundation and a unified terminology. By proposing a holistic and integrated solution across multiple stakeholders, the model emphasizes quality and the extensive impact of championing accountability, sustainability, and responsible practices with global traceability. With these combined efforts, the food supply chain moves toward a global tracking and tracing process that not only ensures product quality but also addresses its impact on a broader scale, fostering accountability, sustainability, and responsible food production and consumption. © 2025, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DRL-Based Time-Varying Workload Scheduling With Priority and Resource Awareness

引用

IEEE Transactions on network and Service Management 2025年

作者： Liu, Qifeng Fan, Qilin Zhang, Xu Li, Xiuhua Wang, Kai Xiong, Qingyu Chongqing University School of Big Data and Software Engineering Chongqing400044 China Chongqing University Key Laboratory of Dependable Service Computing in Cyber Physical Society of Ministry of Education Chongqing400044 China Nanjing University School of Electronic Science and Engineering Nanjing210023 China Haihe Laboratory of Information Technology Application Innovation Tianjin300072 China Harbin Institute of Technology School of Computer Science and Technology Weihai264209 China Shandong Key Laboratory of Industrial Network Security Weihai264209 China

With the proliferation of cloud services and the continuous growth in enterprises' demand for dynamic multi-dimensional resources, the implementation of effective strategy for time-varying workload scheduling has become increasingly significant. In this paper, we propose a deep reinforcement learning (DRL)-based method for time-varying workload scheduling, aiming to allocate resources efficiently across servers in the cluster. Specifically, we integrate a classifier and queue scorer to construct a priority queue that exploits temporal resource utilization patterns across different workload classes. Then, we design parallel graph attention layers to capture the dimensional features and temporal dynamics of cloud server cluster. Moreover, we propose a DRL algorithm to generate scheduling strategies that can adapt to dynamic environments. Validation on real-world traces from Google cluster demonstrates that our method outperforms existing approaches in key metrics of cloud server cluster management. © 2004-2012 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Multimodal Dialogue Emotion Recognition Based on Label Optimization and Coarse-Grained Assisted Fine-Grained

Multimodal Dialogue Emotion Recognition Based on Label Optim...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Zhibo Zhang Zhenyu Yang Yan Huang Baojie Xu Wenyue Hu Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) China Shandong Provincial Key Laboratory of Industrial Network and Information System Security Shandong Fundamental Research Center for Computer Science China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Multimodal dialogue emotion recognition integrates data from multiple modalities to accurately identify emotional states in conversations. However, differences in expression and information density across modalities complicate the fusion of features. Traditional methods may introduce redundant information from other utterances, reducing the accuracy of emotion recognition. Existing one-hot labels often fail to capture the full range of emotional expressions, leading to biased results. To address these issues, we propose a model that fuses different modalities within the same utterance to avoid redundancy. It employs a progressive classification process, refining emotion recognition from coarse to fine granularity. Additionally, we use emotion polarity probabilities as weights for fine-grained classification and introduce a multimodal information-rich label that considers both the data and their interactions. Experiments on IEMOCAP and MELD datasets demonstrate the model’s effectiveness, significantly improving dialog emotion recognition accuracy. Our code is available at https://***/r/LOCG-188E.

关键词： Emotion recognition Accuracy Fuses Refining Termination of employment Speech recognition Signal processing Speech processing Optimization Faces

来源：评论

学校读者我要写书评

暂无评论

A Time-Driven UAV Deployment and Content Caching Approach for Multi-UAV-Enabled Consumer Electronics network

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Zhang, Jianshan Yang, Xu Ming, Rui Luo, Haibo Wang, Xiaoding Piran, Md. Jalil Minjiang University School of Computer and Big Data Fuzhou350108 China Fujian Normal University College of Computer and Cyber Security Fujian Provincial Key Laboratory of Network Security and Cryptology Fujian Fuzhou350117 China Chinese Academy of Tropical Agricultural Sciences National Key Laboratory for Tropical Crop Breeding Institute of Tropical Bioscience and Biotechnology Sanya Research Institute Hainan Sanya572024 China Sejong University Department of Computer Science and Engineering Seoul05006 Korea Republic of

To address the problem of high delay due to the explosive growth of mobile data traffic in consumer electronics network (CEN), in this paper, we investigate the multiple-unmanned aerial vehicle (UAV)-enabled CEN content caching problem, focusing on minimizing content access delay by jointly optimizing UAV deployment, ground user (GU) access, and content caching while satisfying other resource constraints. We propose an algorithm based on alternating optimization for the scenario where multiple UAVs and base station (BS) collaborate to provide caching services. The algorithm attains a sub-optimal solution with a convergence guarantee by iteratively solving the three sub-problems derived from the decomposition of the target problem. First, a convex difference-based algorithm solves the UAV deployment sub-problem. Second, a greedy-based algorithm is used to solve the content caching sub-problem. Then, a continuous convex approximation algorithm based on the penalty function is used to solve the user access sub-problem. Finally, the above procedure is repeated to obtain a sub-optimal solution to the target problem. The proposed joint optimization algorithm is shown to outperform the benchmark algorithms in terms of average content access delay and cache hit ratio, as evidenced by numerical results. © 1975-2011 IEEE.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

Threat Model and Defense Scheme for Side-Channel Attacks in Client-Side Deduplication

引用

Tsinghua Science and Technology 2023年第1期28卷 1-12页

作者： Guanxiong Ha Hang Chen Chunfu Jia Mingyue Li College of Cyber Science Nankai UniversityTianjin 300350China Tianjin Key Laboratory of Network and Data Security Technology Tianjin 300350China

In cloud storage,client-side deduplication is widely used to reduce storage and communication *** client-side deduplication,if the cloud server detects that the user’s outsourced data have been stored,then clients will not need to reupload the ***,the information on whether data need to be uploaded can be used as a side-channel,which can consequently be exploited by adversaries to compromise data *** this paper,we propose a new threat model against side-channel *** from existing schemes,the adversary could learn the approximate ratio of stored chunks to unstored chunks in outsourced files,and this ratio will affect the probability that the adversary compromises the data privacy through side-channel *** this threat model,we design two defense schemes to minimize privacy leakage,both of which design interaction protocols between clients and the server during deduplication checks to reduce the probability that the adversary compromises data *** analyze the security of our schemes,and evaluate their performances based on a real-world *** with existing schemes,our schemes can better mitigate data privacy leakage and have a slightly lower communication cost.

关键词： cloud storage deduplication side-channel privacy

来源：评论

学校读者我要写书评

暂无评论

Bidirectionally Guided Large Language Models for Consumer-Centric Personalized Recommendation

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Yu, Linfang Xiao, Peng Xu, Liqun Kumari, Saru Alenazi, Mohammed J.F. Industrial Research Institute Chengdu610213 China University of Electronic Science and Technology of China Network and Data Security Key Laboratory of Sichuan Province Chengdu610054 China Chengdu University of Technology School of Computer and Network Security Chengdu China Industrial Research Institute Chengdu610213 China Chaudhary Charan Singh University Department of Mathematics Uttar Pradesh Meerut250004 India Department of Computer Engineering Riyadh11451 Saudi Arabia

With the increasing abundance of consumer electronics, consumers are facing the challenge of information overload, making personalized recommendation systems (RSs) crucial for enhancing consumer experience and promoting sales. Recently, LLM-driven RSs have attracted widespread attention. Nevertheless, the application of LLMs in recommendation still encounters challenges: how to effectively integrate user/item content (CT) and collaborative filtering (CF) information;how to effectively process heterogeneous token sequences;and how to efficiently generate recommendations. Therefore, we propose BGLLM4Rec, a bidirectionally guided LLM for recommendation. Specifically, a bidirectionally guided pre-training strategy is proposed to extend user/item ID tokens into the vocabulary of the pre-trained LLM, and leverage both CF LLM and CT LLM to learn their embedding representations. Meanwhile, a heterogeneous-homogeneous (Het-Hom) prompting strategy is proposed to decompose the input sequence into heterogeneous prompt and homogeneous main text parts, performing language modeling solely on the latter to enhance modeling effectiveness and stability. Finally, a recommendation-specific masked prompting fine-tuning strategy is employed to further refine the pre-trained CF LLM, enabling the efficient generation of multiple recommended items without hallucinations. Extensive experiments conducted on three public datasets have demonstrated that BGLLM4Rec exhibits exceptional performance in recommendation tasks. © 1975-2011 IEEE.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

PyionNet: Pyramid Progressive Cross-Fusion network for Joint Classification of Hyperspectral and LiDAR data

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 12042-12058页

作者： Pan, Haizhu Zhang, Quanxiu Ge, Haimiao Liu, Moqi Shi, Cuiping Qiqihar University College of Computer and Control Engineering Qiqihar161000 China Qiqihar University Heilongjiang Key Laboratory of Big Data Network Security Detection and Analysis Qiqihar161000 China Xi'an Jiaotong University School of Information and Communications Engineering Shanxi Xi'an710049 China Huzhou University College of Information Engineering Huzhou313000 China

Joint hyperspectral (HS) and Light Detection and Ranging (LiDAR) data for land cover classification have received widespread attention. Nevertheless, existing methods often struggle with multi-scale feature extraction and multi-source feature fusion. To address these issues, this paper proposes a pyramid progressive cross-fusion network, named PyionNet, for joint classification of HS and LiDAR data. Technically, the proposed PyionNet incorporates a pyramid multi-scale extraction (PME) module cascaded with a progressive cross-fusion (PCF) module, which are designed to extract multi-scale features and fuse multi-source features, respectively. Specifically, the PME module employs pyramid spectral and spatial convolutional blocks to extract and aggregate features from HS and LiDAR data, respectively, then uses an expansion-compression strategy to eliminate redundant information, and thus achieve a compact multi-scale feature representation. Moreover, the PCF module utilizes the transformer architecture and introduces two innovative feature fusion blocks, i.e., the discrepancy feature fusion (DFF) block and the common feature fusion (CFF) block, to progressively integrate multi-source and multi-scale features extracted by the PME module. The DFF block is designed to integrate discrepancy features between HS and LiDAR data to enhance complementary information. Subsequently, the CFF block focuses on fusing common features to progressively enrich the fused representation of the DFF block and further enhance the representation power. Experiments on three real-world datasets demonstrate that the proposed PyionNet outperforms state-of-the-art methods in both qualitative and quantitative results. © 2008-2012 IEEE.

关键词： Feature Selection

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：