检索结果-内蒙古大学图书馆

FedCAPrivacy: Privacy-Preserving Heterogeneous Federated Learning with Anonymous Adaptive Clustering

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wei, Yunan Zhao, Shengnan Zhao, Chuan Liu, Zhe Chen, Zhenxiang Zhao, Minghao Shandong Key Laboratory of Ubiquitous Intelligent Computing University of Jinan Jinan250022 China Quan Cheng Laboratory Jinan250103 China Zhejiang Laboratory Hangzhou311121 China School of Data Science and Engineering East China Normal University Shanghai200050 China

Federated learning (FL) is a distributed machine learning paradigm enabling multiple clients to train a model collaboratively without exposing their local data. Among FL schemes, clustering is an effective technique addressing the heterogeneity issue (i.e., differences in data distribution and computational ability affect training performance and effectiveness) via grouping participants with similar computational resources or data distribution into clusters. However, intra-cluster data exchange poses privacy risks, while cluster selection and adaptation introduce challenges that may affect overall performance. To address these challenges, this paper introduces anonymous adaptive clustering, a novel approach that simultaneously enhances privacy protection and boosts training efficiency. Specifically, an oblivious shuffle-based anonymization method is designed to safeguard user identities and prevent the aggregation server from inferring similarities through clustering. Additionally, to improve performance, we introduce an iteration-based adaptive frequency decay strategy, which leverages variability in clustering probabilities to optimize training dynamics. With these techniques, we build the FedCAPrivacy;experiments show that FedCAPrivacy achieves ∼ 7× improvement in terms of performance while maintaining high privacy. Copyright © 2025, The Authors. All rights reserved.

关键词： Federated learning

A neural kernel framework for relation extraction

学校读者我要写书评

暂无评论

Expert Systems with Applications 2025年 286卷

作者： Wang, Kai Chen, Yanping Yang, Weizhe Huang, Ruizhang Qin, Yongbin Text Computing & Cognitive Intelligence Engineering Research Center of National Education Ministry College of Computer Science and Technology Guizhou University China State Key Laboratory of Public Big Data Guizhou University China

Relation extraction (RE) poses significant challenges due to the complexity of identifying semantic relationships between overlapping entity pairs within sentences. Traditional kernel-based methods effectively leverage instance-level similarities but heavily depend on manually designed kernel functions, requiring substantial domain expertise and limiting their flexibility. On the other hand, neural network approaches excel at automatically learning abstract feature representations but typically neglect critical instance-specific information, potentially missing valuable relational details during inference. These limitations motivate the exploration of a hybrid approach that effectively integrates the strengths of traditional kernels with the representational power of neural networks. In this paper, we propose a neural kernel framework designed to bridge this critical gap. The proposed framework employs neural kernels, whose parameters are initialized based on task-specific objectives and optimized through neural training procedures, enabling adaptive learning of similarity measures without relying on handcrafted kernel functions. By maintaining instance-level details through annotated training examples, neural kernels create discriminative, flexible decision boundaries tailored specifically to the relation extraction task. We further develop three complementary neural kernel components, instance kernel, description kernel, and cluster kernel, to show the advantages of kernel substitution. Extensive experiments on the ACE 2005, SemEval 2010, and CoNLL 2004 datasets demonstrate that our neural kernel framework significantly outperforms existing state-of-the-art methods, achieving F1-scores of 88.11 %, 91.08 %, and 98.63 %, respectively. © 2025 Elsevier Ltd

关键词： Natural language processing systems

DULDA: Dual-domain Unsupervised Learned Descent Algorithm for PET image reconstruction

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Hu, Rui Chen, Yunmen Kim, Kyungsang Rockenbach, Marcio Aloisio Bezerra Cavalcanti Li, Quanzheng Liu, Huafeng State Key Laboratory of Modern Optical Instrumentation Department of Optical Engineering Zhejiang University Hangzhou310027 China Department of Mathematics University of Florida GainesvilleFL32611 United States The Center for Advanced Medical Computing and Analysis Massachusetts General Hospital Harvard Medical School BostonMA02114 United States Center for Clinical Data Science Massachusetts General Brigham BostonMA02116 United States

Deep learning based PET image reconstruction methods have achieved promising results recently. However, most of these methods follow a supervised learning paradigm, which rely heavily on the availability of high-quality training labels. In particular, the long scanning time required and high radiation exposure associated with PET scans make obtaining these labels impractical. In this paper, we propose a dual-domain unsupervised PET image reconstruction method based on learned descent algorithm, which reconstructs high-quality PET images from sinograms without the need for image labels. Specifically, we unroll the proximal gradient method with a learnable l2,1 norm for PET image reconstruction problem. The training is unsupervised, using measurement domain loss based on deep image prior as well as image domain loss based on rotation equivariance property. The experimental results demonstrate the superior performance of proposed method compared with maximum-likelihood expectation–maximization (MLEM), total-variation regularized EM (EM-TV) and deep image prior based method (DIP). Copyright © 2023, The Authors. All rights reserved.

关键词： Image reconstruction

Robust Optimization for Quantum Reinforcement Learning Control using Partial Observations

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Jiang, Chen Pan, Yu Wu, Zheng-Guang Gao, Qing Dong, Daoyi State Key Laboratory of Industrial Control Technology Institute of Cyber-Systems and Control College of Control Science and Engineering Zhejiang University Hangzhou310027 China The School of Automation Science and Electrical Engineering Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing100191 China The School of Engineering and Information Technology University of New South Wales CanberraACT2600 Australia

The current quantum reinforcement learning control models often assume that the quantum states are known a priori for control optimization. However, full observation of quantum state is experimentally infeasible due to the exponential scaling of the number of required quantum measurements on the number of qubits. In this paper, we investigate a robust reinforcement learning method using partial observations to overcome this difficulty. This control scheme is compatible with near-term quantum devices, where the noise is prevalent and predetermining the dynamics of quantum state is practically impossible. We show that this simplified control scheme can achieve similar or even better performance when compared to the conventional methods relying on full observation. We demonstrate the effectiveness of this scheme on examples of quantum state control and quantum approximate optimization algorithm. It has been shown that high-fidelity state control can be achieved even if the noise amplitude is at the same level as the control amplitude. Besides, an acceptable level of optimization accuracy can be achieved for QAOA with noisy control Hamiltonian. This robust control optimization model can be trained to compensate the uncertainties in practical quantum computing. © 2022, CC BY.

关键词： Robust control

Demystifying the vulnerability propagation and its evolution via dependency trees in the NPM ecosystem

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Liu, Chengwei Chen, Sen Fan, Lingling Chen, Bihuan Liu, Yang Peng, Xin College of Intelligence and Computing Tianjin University Tianjin China College of Cyber Science Nankai University Tianjin China School of Computer Science Shanghai Key Laboratory of Data Science Fudan University Shanghai China School of Computer Science and Engineering Nanyang Technological University Singapore Singapore Nanyang Technological University Singapore

Third-party libraries with rich functionalities facilitate the fast development of JavaScript software, leading to the explosive growth of the NPM ecosystem. However, it also brings new security threats that vulnerabilities could be introduced through dependencies from third-party libraries. In particular, the threats could be excessively amplified by transitive dependencies. Existing research only considers direct dependencies or reasoning transitive dependencies based on reachability analysis, which neglects the NPM-specific dependency resolution rules as adapted during real installation, resulting in wrongly resolved dependencies. Consequently, further fine-grained analysis, such as precise vulnerability propagation and their evolution over time in dependencies, cannot be carried out precisely at a large scale, as well as deriving ecosystem-wide solutions for vulnerabilities in dependencies. To fill this gap, we propose a knowledge graph-based dependency resolution, which resolves the inner dependency relations of dependencies as trees (i.e., dependency trees), and investigates the security threats from vulnerabilities in dependency trees at a large scale. Specifically, we first construct a complete dependencyvulnerability knowledge graph (DVGraph) that captures the whole NPM ecosystem (over 10 million library versions and 60 million well-resolved dependency relations). Based on it, we propose a novel algorithm (DTResolver) to statically and precisely resolve dependency trees, as well as transitive vulnerability propagation paths, for each package by taking the official dependency resolution rules into account. Based on that, we carry out an ecosystem-wide empirical study on vulnerability propagation and its evolution in dependency trees. Our study unveils lots of useful findings, and we further discuss the lessons learned and solutions for different stakeholders to mitigate the vulnerability impact in NPM based on our findings. For example, we implement a depend

关键词： Libraries

Social-enhanced recommendation using graph-based contrastive learning

学校读者我要写书评

暂无评论

Social-enhanced recommendation using graph-based contrastive...

IEEE International Conference on High Performance computing and Communications (HPCC)

作者： Peng Xue Qian Gao Jun Fan Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan Shandong China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan Shandong China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan Shandong China China Telecom Digital Intelligence Techonology Co Ltd Jinan Shandong China

The social network-based recommendation model use social network information to mitigate data sparsity issues and improve the accuracy of recommendation models. However, In most social network-based recommendation algorithms, the social neighbors' contributions are difficult to distinguish from the central user's, neglecting hidden correlations in social infor-mation. In order to solve this problem, this paper presents a graph-based contrastive learning framework for social-enhanced recommendation (SoGCLR) that utilizes implicit social information captured by a new social relation attention mechanism, en-rich user representations, and improve model robustness through graph-based contrastive learning. Specifically, the paper captures the degree of contribution of each neighbor in the social graph to the central user through a social relation attention layer, thus obtaining hidden correlations in social information, and further integrates this with user information in the user-item interaction graph to enrich user representations. In addition, the paper incorporates graph-based contrastive learning into the recommendation task using cross-layer contrastive learning, calculating contrastive loss and mapping nodes with similar but different exposure rates to nearby regions to mitigate exposure bias issues. Results from Ciao and Epinions demonstrate that SoGCLR reduces RMSE and MAE by 1.33% to 1.84% compared with baseline models.

关键词：

P-FedAvg: Parallelizing federated learning with theoretical guarantees 40

学校读者我要写书评

暂无评论

P-FedAvg: Parallelizing federated learning with theoretical ...

40th IEEE Conference on Computer Communications, INFOCOM 2021

作者： Zhong, Zhicong Zhou, Yipeng Wu, Di Chen, Xu Chen, Min Li, Chao Sheng, Quan Z. Sun Yat-Sen University School of Computer Science and Engineering Guangzhou510006 China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou510006 China PCL Research Center of Networks and Communications Peng Cheng Laboratory Shenzhen518000 China Department of Computing Macquarie University SydneyNSW2109 Australia School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China Tencent Technology Co. Ltd Shenzhen China

ISBN: (纸本)9780738112817

With the growth of participating clients, the centralized parameter server (PS) will seriously limit the scale and efficiency of Federated Learning (FL). A straightforward approach to scale up the FL system is to construct a Parallel FL (PFL) system with multiple PSes. However, it is unclear whether PFL can really achieve a faster convergence rate or not. Even if the answer is yes, it is non-trivial to design a highly efficient parameter average algorithm for a PFL system. In this paper, we propose a completely parallelizable FL algorithm called P-FedAvg under the PFL architecture. P-FedAvg extends the well-known FedAvg algorithm by allowing multiple PSes to cooperate and train a learning model together. In P-FedAvg, each PS is only responsible for a fraction of total clients, but PSes can mix model parameters in a dedicatedly designed way so that the FL model can well converge. Different from heuristic-based algorithms, P-FedAvg is with theoretical guarantees. To be rigorous, we conduct theoretical analysis on the convergence rate of P-FedAvg, and derive the optimal weights for each PS to mix parameters with its neighbors. We also examine how the overlay topology formed by PSes affects the convergence rate and robustness of a PFL system. Lastly, we perform extensive experiments with real datasets to verify our analysis and demonstrate that P-FedAvg can significantly improve convergence rates than traditional FedAvg and other competitive baselines. We believe that our work can help to lay a theoretical foundation for building more efficient PFL systems. © 2021 IEEE.

关键词： Network topology

LI-FPN: Depression and Anxiety Detection from Learning and Imitation

学校读者我要写书评

暂无评论

LI-FPN: Depression and Anxiety Detection from Learning and I...

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Xingyun Li Lin Lu Xinyu Yi Hao Wang Yunshao Zheng Yanhong Yu Qingxiang Wang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Shandong Mental Health Center Jinan China Shandong University of Traditional Chinese Medicine Jinan China

With the rise in societal pressures, depression and anxiety have increasingly become prominent mental health conditions impacting people’s lives. To enhance the efficacy of automatic detection for these disorders, we have developed an experimental framework called the Voluntary Facial Expression Mimicry(VFEM). This framework led to the creation of the VFEM dataset, which supports related research endeavors. Subsequently, we introduce the LI-FPN designed specifically for the automatic identification of depression and anxiety disorders. The LI-FPN comprises two core components: the Learning and Imitation Module(LIM) and the Spatio-temporal Feature Pyramid Network(STFPN). Within the LIM, we leverage sequence features to facilitate comprehensive feature extraction through learning and imitation steps. The STFPN is designed to focus on outliers in multi-scale features for further screening. Compared with traditional attention methods, LI-FPN is more suitable for processing sequence data features and small sample datasets. Upon training using the VFEM dataset, the LI-FPN achieves impressive accuracies: 0.850 for depression detection, 0.835 for anxiety detection, and 0.786 for co-occurrence detection of depression and anxiety. Meanwhile, LI-FPN also achieves SOAT results on AVEC2014 dataset. The source code for LI-FPN is accessible at https://***/muzixingyun/LI-FPN

关键词：

Copula Guided Parallel Gibbs Sampling for Nonparametric and Coherent Topic Discovery (Extended Abstract)

学校读者我要写书评

暂无评论

Copula Guided Parallel Gibbs Sampling for Nonparametric and ...

International Conference on data engineering

作者： Lihui Lin Yanghui Rao Haoran Xie Raymond Y. K. Lau Jian Yin Fu Lee Wang Qing Li School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Department of Computing and Decision Sciences Lingnan University New Territories Hong Kong SAR Department of Information Systems City University of Hong Kong Kowloon Hong Kong SAR Guangdong Key Laboratory of Big Data Analysis and Processing Sun Yat-sen University Guangzhou China School of Science and Technology Hong Kong Metropolitan University Kowloon Hong Kong SAR Department of Computing The Hong Kong Polytechnic University Kowloon Hong Kong SAR

In terms of the generative process, the Gamma-Gamma-Poisson Process (G2PP) is equivalent to the nonparametric topic model of Hierarchical Dirichlet Process (HDP). Considering the high computational cost of estimating parameters in HDP, a parallel G2PP was developed to generate topics efficiently via multi-threading. Unfortunately, the above model needs to predefine the number of topics. To address this issue, we first propose a Topic Self-Adaptive Model (TSAM) for nonparametric and parallel topic discovery. In TSAM, a monitor-executor mechanism is developed to manage the global topic information using a hierarchical structure of threads. Based on the apparatus of copulas, we further extend our TSAM to TSAMcop for coherent topic modeling by exploiting a copula guided parallel Gibbs sampling algorithm. Extensive experiments validate the effectiveness of both TSAM and TSAMcop.

关键词：