检索结果-内蒙古大学图书馆

VPFL:A verifiable privacy-preserving federated learning scheme for edge computing systems

Digital Communications and Networks 2023年第4期9卷 981-989页

作者： Jiale Zhang Yue Liu Di Wu Shuai Lou Bing Chen Shui Yu School of Information Engineering Yangzhou UniversityYangzhou225009China Water Conservancy and Civil Engineering College Inner Mongolia Agricultural UniversityHohhot010018China Deakin Blockchain Innovation Lab School of Information TechnologyDeakin UniversityMelbourneVIC3125Australia College of Computer Science and Technology Nanjing University of Aeronautics and AstronauticsNanjing211106China School of Computer Science University of Technology SydneySydney2007Australia

Federated learning for edge computing is a promising solution in the data booming era,which leverages the computation ability of each edge device to train local models and only shares the model gradients to the central ***,the frequently transmitted local gradients could also leak the participants’private *** protect the privacy of local training data,lots of cryptographic-based Privacy-Preserving Federated Learning(PPFL)schemes have been ***,due to the constrained resource nature of mobile devices and complex cryptographic operations,traditional PPFL schemes fail to provide efficient data confidentiality and lightweight integrity verification *** tackle this problem,we propose a Verifiable Privacypreserving Federated Learning scheme(VPFL)for edge computing systems to prevent local gradients from leaking over the transmission ***,we combine the Distributed Selective Stochastic Gradient Descent(DSSGD)method with Paillier homomorphic cryptosystem to achieve the distributed encryption functionality,so as to reduce the computation cost of the complex ***,we further present an online/offline signature method to realize the lightweight gradients integrity verification,where the offline part can be securely outsourced to the edge *** security analysis demonstrates the proposed VPFL can achieve data confidentiality,authentication,and *** last,we evaluate both communication overhead and computation cost of the proposed VPFL scheme,the experimental results have shown VPFL has low computation costs and communication overheads while maintaining high training accuracy.

关键词： Federated learning Edge computing Privacy-preserving Verifiable aggregation Homomorphic cryptosystem

来源：评论

学校读者我要写书评

暂无评论

Robust Hashing with Deep Features and Meixner Moments for Image Copy Detection

引用

ACM Transactions on Multimedia Computing, Communications and Applications 2024年第12期20卷 1-23页

作者： Yu, Mengzhu Tang, Zhenjun Liang, Xiaoping Zhang, Xianquan Li, Zhixin Zhang, Xinpeng Key Lab of Education Blockchain and Intelligent Technology Ministry of Education Guangxi Key Lab of Multi-Source Information Mining and Security Guangxi Normal University Guilin China School of Computer Science Fudan University Shanghai China

Copy detection is a key task of image copyright protection. Most robust hashing schemes do not make satisfied performance of image copy detection yet. To address this, a robust hashing scheme with deep features and Meixner moments is proposed for image copy detection. In the proposed hashing, global deep features are extracted by applying tensor Singular Value Decomposition (t-SVD) to the three-order tensor constructed in the DWT domain of the feature maps calculated by the pre-trained VGG16. Since the feature maps in the DWT domain are slightly disturbed by digital operations, the constructed three-order tensor is stable and thus the desirable robustness is guaranteed. Moreover, since t-SVD can decompose a three-order tensor into multiple low-dimensional matrices reflecting intrinsic structure, the global deep feature calculation from the low-dimensional matrices can provide good discrimination. Local features are calculated by the block-based Meixner moments. As the Meixner moments are resistant to geometric transformation and can efficiently discriminate various images, the use of the block-based Meixner moments can make discriminative and robust local features. Hash is ultimately determined by quantifying and combining global deep features and local features. The results of extensive experiments on open image datasets demonstrate that the proposed robust hashing outperforms some state-of-the-art robust hashing schemes in terms of classification and copy detection performances. © 2024 Copyright held by the owner/author(s).

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Text-based person search by non-saliency enhancing and dynamic label smoothing

引用

Neural Computing and Applications 2024年第21期36卷 13327-13339页

作者： Pang, Yonghua Zhang, Canlong Li, Zhixin Wei, Chunrong Wang, Zhiwen Key Lab of Education Blockchain and Intelligent Technology Ministry of Education Guangxi Normal University Guilin China Guangxi Key Lab of Multi-source Information Mining and Security Guangxi Normal University Guilin541004 China School of Computer Science and Technology Guangxi University of Science and Technology Liuzhou545006 China

The current text-based person re-identification (re-ID) models tend to learn salient features of image and text, which however is prone to failure in identifying persons with very similar dress, because their image contents with observable but indescribable difference may have identical textual description. To address this problem, we propose a re-ID model based on saliency masking to learn non-salient but highly discriminative features, which can work together with the salient features to provide more robust pedestrian identification. To further improve the performance of the model, a cross-modal projection matching loss with dynamic label smoothing (named CMPM-DS) is proposed to train our model, and our CMPM-DS can adaptively adjust the smoothing degree of the true distribution. We conduct extensive ablation and comparison experiments on two popular re-ID benchmarks to demonstrate the efficiency of our model and loss function, and our model achieves SOTA, improving the existing best R@1 by 0.33% on CUHK-PEDE and 4.45% on RSTPReID. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Person re-identification Cross-modal projection matching Saliency masking label smoothing

来源：评论

学校读者我要写书评

暂无评论

DecTest: A Decentralised Testing Architecture for Improving Data Accuracy of blockchain Oracle

DecTest: A Decentralised Testing Architecture for Improving ...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Zeng, Xueying Xian, Youquan Li, Chunpei Hu, Zhengdong Zhou, Aoxiang Liu, Peng School of Computer Science and Engineering Guangxi Normal University Key Lab of Education Blockchain and Intelligent Technology Ministry of Education Guilin541004 China

ISBN: (纸本)9781665410205

blockchain technology ensures secure and trust-worthy data flow between multiple participants on the chain, but interoperability of on-chain and off-chain data has always been a difficult problem that needs to be solved. To solve the problem that blockchain systems cannot access off-chain data, oracle is introduced. However, existing research mainly focuses on the consistency and integrity of data, but ignores the problem that oracle nodes may be externally attacked or provide false data for selfish motives, resulting in the unresolved problem of data accuracy. In this paper, we introduce a new Decentralized Testing architecture (DecTest) that aims to improve data accuracy. A blockchain oracle random secret testing mechanism is first proposed to enhance the monitoring and verification of nodes by introducing a dynamic anonymized question-verification committee. Based on this, a comprehensive evaluation incentive mechanism is designed to incentivize honest work performance by evaluating nodes based on their reputation scores. The simulation results show that we successfully reduced the discrete entropy value of the acquired data and the real value of the data by 61.4 %. © 2024 IEEE.

关键词： Data integrity

来源：评论

学校读者我要写书评

暂无评论

SW-Net: A novel few-shot learning approach for disease subtype prediction

引用

BIOCELL 2023年第3期47卷 569-579页

作者： YUHAN JI YONG LIANG ZIYI YANG NING AI Faculty of Innovation Engineering School of Computer Science and EngineeringMacao University of Science and TechnologyMacao999078China Tencent Quantum Lab Shenzhen518000China

Few-shot learning is becoming more and more popular in many fields,especially in the computer vision *** inspires us to introduce few-shot learning to the genomic field,which faces a typical few-shot problem because some tasks only have a limited number of samples with *** goal of this study was to investigate the few-shot disease sub-type prediction problem and identify patient subgroups through training on small *** disease subtype classification allows clinicians to efficiently deliver investigations and interventions in clinical *** propose the SW-Net,which simulates the clinical process of extracting the shared knowledge from a range of interrelated tasks and generalizes it to unseen *** model is built upon a simple baseline,and we modified it for genomic *** initialization for the classifier and transductive fine-tuning techniques were applied in our model to improve prediction accuracy,and an Entropy regularization term on the query set was appended to reduce ***,to address the high dimension and high noise issue,we future extended a feature selection module to adaptively select important features and a sample weighting module to prioritize high-confidence *** on simulated data and The Cancer Genome Atlas meta-dataset show that our new baseline model gets higher prediction accuracy compared to other competing algorithms.

关键词： Few-shot learning Disease sub-type classification Feature selection Deep learning Meta-learning

来源：评论

学校读者我要写书评

暂无评论

Privacy-Preserving Stable Data Trading for Unknown Market Based on blockchain

引用

IEEE Transactions on Mobile Computing 2025年第7期24卷 5615-5631页

作者： Deng, Qingyong Zuo, Qinghua Li, Zhetao Liu, Haolin Xie, Yong Key Lab of Education Blockchain and Intelligent Technology Ministry of Education Guangxi Normal University Guilin541004 China Guangxi Key Lab of Multi-source Information Mining & Security Guangxi Normal University Guilin541004 China College of Information Science and Technology Jinan University Guangzhou510632 China School of Computer Science Xiangtan University Xiangtan Hunan411105 China School of Computer Science Nanjing University of Posts and Telecommunications Nanjing Jiangsu210042 China

Crowdsensing Data Trading (CDT) has emerged as a novel data trading paradigm, where market stability is crucial during the transaction matching process. However, most existing CDT systems usually assume that the preferences of both parties are known and the third-party trading platform is trustworthy, which is impractical in real-world scenarios and leads to significant challenges in reliability and privacy preservation. To address these challenges, we propose a Privacy-Preserving and Stable Data Trading for Unknown Market based on blockchain and Bilateral Reputation (PPSDT-UMBBR) scheme in the decentralized CDT system. Firstly, a privacy-preserving bilateral preference initialization method is designed to achieve the initial matching of buyers and sellers without exposing their location and attribute privacy. Then, a stable matching method based on dynamic bilateral preference updating is proposed, integrating Differential Privacy, Stable matching theory, and a strategy based on Asymmetric Bilateral Preferences with Multi-armed Bandits (DPS-ABPMAB). Finally, we theoretically analyze the security and prove that the market outcome is δ stable. Furthermore, compared to other benchmark methods based on real datasets, our proposed DPS-ABPMAB algorithm improves the average accumulative reward by at least 4.22%, and reduces the average accumulative regret and the mean evaluation error rate by at least 66.86% and 7.35%, respectively. © 2025 IEEE.

关键词： Marketplaces

来源：评论

学校读者我要写书评

暂无评论

A Privacy-Preserving and Efficient Data Sharing Scheme Based on blockchain in IIoT 22

A Privacy-Preserving and Efficient Data Sharing Scheme Based...

引用

22nd IEEE International Symposium on Parallel and Distributed Processing with Applications, ISPA 2024

作者： Peng, Hongyan Yang, Yipeng Li, Dongcheng Wang, Peng Liu, Peng Guangxi Normal University Key Lab of Education Blockchain and Intelligent Technology Ministry of Education China Guangxi Normal University School of Computer Science and Engineering China

ISBN: (纸本)9798331509712

In the industrial Internet of Things (IIoT) scenarios, data sharing can promote mutual collaboration among production parties to improve productivity and optimise resource allocation, but data sharing in industrial scenarios faces the risk of privacy leakage due to open networks. Attribute-based encryption (ABE) can be used to solve the problem of data sharing privacy leakage. However, there is still no effective privacy protection solution for data sharing and access control policies in blockchain, which may expose sensitive information of data owners. Moreover, existing schemes only focus on protecting data privacy and ignore the protection of users' query privacy. In addition, most of these schemes are based on cloud servers, which may lead to data tampering and a single point of failure. To address these issues, we propose a blockchain-based data security sharing scheme (BAPIR) that combines attribute-based encryption and private information retrieval (PIR). In this paper, we implement fine-grained access control using ABE based on the inner product and protecting the access policy. Additionally, PIR technology is introduced to protect users' query privacy. To achieve efficient blockchain data storage, IPFS is used for on-chain and off-chain collaboration, and complex computational tasks are outsourced to edge servers, ensuring secure and efficient data sharing. Finally, we demonstrate the security and efficiency of BAPIR through security analysis and performance evaluation. © 2024 IEEE.

关键词： Information leakage

来源：评论

学校读者我要写书评

暂无评论

Dynamic Graph Information Bottleneck 24

Dynamic Graph Information Bottleneck

引用

33rd ACM Web Conference, WWW 2024

作者： Yuan, Haonan Sun, Qingyun Fu, Xingcheng Ji, Cheng Li, Jianxin School of Computer Science and Engineering BDBC Beihang University Beijing China Key Lab of Education Blockchain and Intelligent Technology Guangxi Normal University Guangxi Guilin China

ISBN: (纸本)9798400701719

Dynamic Graphs widely exist in the real world, which carry complicated spatial and temporal feature patterns, challenging their representation learning. Dynamic Graph Neural Networks (DGNNs) have shown impressive predictive abilities by exploiting the intrinsic dynamics. However, DGNNs exhibit limited robustness, prone to adversarial attacks. This paper presents the novelDynamic Graph Information Bottleneck (DGIB) framework to learn robust and discriminative representations. Leveraged by the Information Bottleneck (IB) principle, we first propose the expected optimal representations should satisfy theMinimal-Sufficient-Consensual (MSC) Condition. To compress redundant as well as conserve meritorious information into latent representation, DGIB iteratively directs and refines the structural and feature information flow passing through graph snapshots. To meet theMSC Condition, we decompose the overall IB objectives into DGIBMS and DGIBC, in which the DGIB_MS channel aims to learn the minimal and sufficient representations, with the DGIBC channel guarantees the predictive consensus. Extensive experiments on real-world and synthetic dynamic graph datasets demonstrate the superior robustness of DGIB against adversarial attacks compared with state-of-the-art baselines in the link prediction task. To the best of our knowledge, DGIB is the first work to learn robust representations of dynamic graphs grounded in the information-theoretic IB principle. © 2024 Owner/Author.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

MUNE: A Multi-Level User-News Embedding Model for Personalised News Recommendations 24

MUNE: A Multi-Level User-News Embedding Model for Personalis...

引用

8th International Conference on computer science and Artificial Intelligence, CSAI 2024

作者： Long, Haixiang Luo, Xudong School of Computer Science and Engineering Guangxi Key Lab of Multi-Source Information Mining & Security Key Lab of Education Blockchain and Intelligent Technology of Ministry of Education Guangxi Normal University Guangxi Guilin China

ISBN: (纸本)9798400718182

Personalised news recommendation systems are critical for delivering relevant content, but effectively capturing and modelling users’ diverse and evolving interests is challenging. In this paper, we introduce a Multi-level User-News Embedding (MUNE) model for personalised news recommendations. MUNE uses pre-trained BERT embeddings to generate contextual representations, enabling dynamic, multi-level interaction modelling between users and candidate news. It incorporates a Dynamic Multi-Level Interaction Attention (DMIA) module for capturing complex interactions and a Dynamic Similarity Unit (DSU) for enhancing feature fusion using cosine similarity and non-linear transformations. Evaluated on large-scale datasets (MIND and EB-NeRD), MUNE significantly outperforms state-of-the-art baselines in key metrics, with ablation studies validating the importance of DMIA and DSU in improving user interest modelling and recommendation quality. MUNE’s integration of multi-level features makes it highly effective in capturing user behaviours and understanding news content. © 2024 Copyright held by the owner/author(s).

关键词： Linear transformations

来源：评论

学校读者我要写书评

暂无评论

Robust Graph Learning Against Adversarial Evasion Attacks via Prior-Free Diffusion-Based Structure Purification 25

Robust Graph Learning Against Adversarial Evasion Attacks vi...

引用

34th ACM Web Conference, WWW 2025

作者： Luo, Jiayi Sun, Qingyun Yuan, Haonan Fu, Xingcheng Li, Jianxin SKLCCSE School of Computer Science and Engineering Beihang University Beijing China Key Lab of Education Blockchain and Intelligent Technology Guangxi Normal University Guangxi China

ISBN: (纸本)9798400712746

Adversarial evasion attacks pose significant threats to graph learning, with lines of studies that have improved the robustness of Graph Neural Networks (GNNs). However, existing works rely on priors about clean graphs or attacking strategies, which are often heuristic and inconsistent. To achieve robust graph learning over different types of evasion attacks and diverse datasets, we investigate this problem from a prior-free structure purification perspective. Specifically, we propose a novel Diffusion-based Structure Purification framework named DiffSP, which creatively incorporates the graph diffusion model to learn intrinsic distributions of clean graphs and purify the perturbed structures by removing adversaries under the direction of the captured predictive patterns without relying on priors. DiffSP is divided into the forward diffusion process and the reverse denoising process, during which structure purification is achieved. To avoid valuable information loss during the forward process, we propose an LID-driven non-isotropic diffusion mechanism to selectively inject noise anisotropically. To promote semantic alignment between the clean graph and the purified graph generated during the reverse process, we reduce the generation uncertainty by the proposed graph transfer entropy guided denoising mechanism. Extensive experiments demonstrate the superior robustness of DiffSP against evasion attacks. © 2025 Copyright held by the owner/author(s).

关键词： Graph neural networks

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：