Byzantine-Resilient Federated Machine Learning via Over-the-Air Computation

doi:10.1109/ICCWorkshops50388.2021.9473694

	Byzantine-Resilient Federated Machine Learning via Over-the-Air Computation
	Huang, Shaoming1,2,3 ; Zhou, Yong1 ; Wang, Ting 4; Shi, Yuanming1
	2021-06-01
会议录名称	2021 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS WORKSHOPS, ICC WORKSHOPS 2021 - PROCEEDINGS
ISSN	2164-7038
发表状态	已发表
DOI	10.1109/ICCWorkshops50388.2021.9473694
摘要	Federated learning (FL) is recognized as a key enabling technology to provide intelligent services for future wireless networks and industrial systems with delay and privacy guarantees. However, the performance of wireless FL can be significantly degraded by Byzantine attack, such as data poisoning attack, model poisoning attack and free-riding attack. To design the Byzantine-resilient FL paradigm in wireless networks with limited radio resources, we propose a novel communication-efficient robust model aggregation scheme via over-the-air computation (AirComp). This is achieved by applying the Weiszfeld algorithm to obtain the smoothed geometric median aggregation against Byzantine attack. The additive structure of the Weiszfeld algorithm is further leveraged to match the signal superposition property of multiple-access channels via AirComp, thereby expediting the communication-efficient secure aggregation process of FL. Numerical results demonstrate the robustness against Byzantine devices and good learning performance of the proposed approach. © 2021 IEEE.
会议录编者/会议主办者	IEEE Communication Society ; IEEE Montreal Section ; IEEE Ottawa Section
关键词	Machine learning Wireless networks Enabling technologies Future wireless networks Intelligent Services Learning performance Multiple access channels Secure aggregations Signal superposition Weiszfeld algorithms
会议名称	2021 IEEE International Conference on Communications Workshops, ICC Workshops 2021
出版地	345 E 47TH ST, NEW YORK, NY 10017 USA
会议地点	Virtual, Online
会议日期	June 14, 2021 - June 23, 2021
URL	查看原文
收录类别	EI ; CPCI-S ; CPCI
语种	英语
资助项目	National Natural Science Foundation of China (NSFC)[62001294]
WOS研究方向	Computer Science ; Engineering ; Telecommunications
WOS类目	Computer Science, Information Systems ; Engineering, Electrical & Electronic ; Telecommunications
WOS记录号	WOS:000848412200177
出版者	Institute of Electrical and Electronics Engineers Inc.
EI入藏号	20213410796265
EI主题词	Privacy by design
EI分类号	716.3 Radio Systems and Equipment
原始文献类型	Conference article (CA)
文献类型	会议论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/133474
专题	信息科学与技术学院_硕士生信息科学与技术学院_PI研究组_石远明组信息科学与技术学院_PI研究组_周勇组
通讯作者	Huang, Shaoming
作者单位	1.ShanghaiTech Univ, Sch Informat Sci & Technol, Shanghai 201210, Peoples R China 2.Chinese Acad Sci, Shanghai Inst Microsyst & Informat Technol, Beijing, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 4.East China Normal Univ, Sch Software Engn, Shanghai Key Lab Trustworthy Comp, Shanghai, Peoples R China
第一作者单位	信息科学与技术学院
通讯作者单位	信息科学与技术学院
第一作者的第一单位	信息科学与技术学院
推荐引用方式 GB/T 7714	Huang, Shaoming,Zhou, Yong,Wang, Ting,et al. Byzantine-Resilient Federated Machine Learning via Over-the-Air Computation[C]//IEEE Communication Society, IEEE Montreal Section, IEEE Ottawa Section. 345 E 47TH ST, NEW YORK, NY 10017 USA:Institute of Electrical and Electronics Engineers Inc.,2021.