Reliable and Privacy-Preserving Federated Learning with Anomalous Users  

作　　者：ZHANG Weiting LIANG Haotian XU Yuhua ZHANG Chuan 

机构地区：[1]Beijing Jiaotong University,Beijing 100091,China [2]Beijing Institute of Technology,Beijing 100081,China

出　　处：《ZTE Communications》2023年第1期15-24,共10页中兴通讯技术（英文版）

基　　金：supported in part by the Fundamental Research Funds for Central Universities under Grant No.2022RC006;in part by the National Nat⁃ural Science Foundation of China under Grant Nos.62201029 and 62202051;in part by the BIT Research and Innovation Promoting Project under Grant No.2022YCXZ031;in part by the Shandong Provincial Key Research and Development Program under Grant No.2021CXGC010106;in part by the China Postdoctoral Science Foundation under Grant Nos.2021M700435,2021TQ0042,2021TQ0041,BX20220029 and 2022M710007.

摘　　要：Recently,various privacy-preserving schemes have been proposed to resolve privacy issues in federated learning(FL).However,most of them ignore the fact that anomalous users holding low-quality data may reduce the accuracy of trained models.Although some existing works manage to solve this problem,they either lack privacy protection for users’sensitive information or introduce a two-cloud model that is difficult to find in reality.A reliable and privacy-preserving FL scheme named reliable and privacy-preserving federated learning(RPPFL)based on a single-cloud model is proposed.Specifically,inspired by the truth discovery technique,we design an approach to identify the user’s reliability and thereby decrease the impact of anomalous users.In addition,an additively homomorphic cryptosystem is utilized to provide comprehensive privacy preservation(user’s local gradient privacy and reliability privacy).We give rigorous theoretical analysis to show the security of RPPFL.Based on open datasets,we conduct extensive experiments to demonstrate that RPPEL compares favorably with existing works in terms of efficiency and accuracy.

关 键 词：federated learning anomalous user privacy preservation reliability homomorphic cryptosystem 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP309[自动化与计算机技术—控制科学与工程]