Byzantine-Resilient Federated Learning at Edge

doi:10.1109/TC.2023.3257510

CORC > 自动化研究所 > 中国科学院自动化研究所 > 数字内容技术与服务研究中心 > 听觉模型与认知计算

	Byzantine-Resilient Federated Learning at Edge
	Tao, Youming 1; Cui, Sijia 4; Xu, Wenlu 3; Yin, Haofei 1; Yu, Dongxiao 1; Liang, Weifa 2; Cheng, Xiuzhen 1
刊名	IEEE TRANSACTIONS ON COMPUTERS
	2023-09-01
卷号	72 期号:9 页码:2600-2614
关键词	Byzantine resilience communication efficiency edge intelligent systems federated learning
ISSN号	0018-9340
DOI	10.1109/TC.2023.3257510
通讯作者	Yu, Dongxiao(dxyu@sdu.edu.cn)
英文摘要	Both Byzantine resilience and communication efficiency have attracted tremendous attention recently for their significance in edge federated learning. However, most existing algorithms may fail when dealing with real-world irregular data that behaves in a heavy-tailed manner. To address this issue, we study the stochastic convex and non-convex optimization problem for federated learning at edge and show how to handle heavy-tailed data while retaining the Byzantine resilience, communication efficiency and the optimal statistical error rates simultaneously. Specifically, we first present a Byzantine-resilient distributed gradient descent algorithm that can handle the heavy-tailed data and meanwhile converge under the standard assumptions. To reduce the communication overhead, we further propose another algorithm that incorporates gradient compression techniques to save communication costs during the learning process. Theoretical analysis shows that our algorithms achieve order-optimal statistical error rate in presence of Byzantine devices. Finally, we conduct extensive experiments on both synthetic and real-world datasets to verify the efficacy of our algorithms.
资助项目	National Key Research and Development Program of China[2020YFB1005900] ; National Natural Science Foundation of China (NSFC)[62122042]
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE COMPUTER SOC
WOS记录号	WOS:001047175700014
资助机构	National Key Research and Development Program of China ; National Natural Science Foundation of China (NSFC)
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/53981]
专题	数字内容技术与服务研究中心_听觉模型与认知计算
通讯作者	Yu, Dongxiao
作者单位	1.Shandong Univ, Sch Comp Sci & Technol, Qingdao 250100, Shandong, Peoples R China 2.City Univ Hong Kong, Dept Comp Sci, Hong Kong, Peoples R China 3.Univ Calif Los Angeles, Dept Stat, Los Angeles, CA 90095 USA 4.Univ Chinese Acad Sci, Inst Automat, Beijing 101408, Peoples R China
推荐引用方式 GB/T 7714	Tao, Youming,Cui, Sijia,Xu, Wenlu,et al. Byzantine-Resilient Federated Learning at Edge[J]. IEEE TRANSACTIONS ON COMPUTERS,2023,72(9):2600-2614.
APA	Tao, Youming.,Cui, Sijia.,Xu, Wenlu.,Yin, Haofei.,Yu, Dongxiao.,...&Cheng, Xiuzhen.(2023).Byzantine-Resilient Federated Learning at Edge.IEEE TRANSACTIONS ON COMPUTERS,72(9),2600-2614.
MLA	Tao, Youming,et al."Byzantine-Resilient Federated Learning at Edge".IEEE TRANSACTIONS ON COMPUTERS 72.9(2023):2600-2614.