Adaptive Federated Learning With Non-IID Data

doi:10.1093/comjnl/bxac118

	Adaptive Federated Learning With Non-IID Data
	Zeng, Yan 1,2,3; Mu, Yuankai 4; Yuan, Junfeng 1; Teng, Siyuan 1; Zhang, Jilin 1,2,3; Wan, Jian 1,2,3; Ren, Yongjian 1,2,3; Zhang, Yunquan 5
	2022-09-30
发表期刊	COMPUTER JOURNAL
ISSN	0010-4620
页码	15
摘要	With the widespread use of Internet of things(IoT) devices, it generates an enormous volume of data, and it is a challenge to mine the IoT data value while ensuring security and privacy. Federated learning is a decentralized approach for training data located on edge devices, such as mobile phones and IoT devices, while keeping privacy, efficiency, and security. However, the Non-IID (non-independent and identically distributed) data, always greatly impacts the performance of the global model. In this paper, we propose a FedDynamic algorithm to solve the statistical challenge of federated learning caused by Non-IID. As Non-IID data can lead to significant differences in model parameters between edge devices, we set different weights for different devices during model aggregation to get a high-performance global model. We analyze and exact key indices (local model accuracy, local data quality, and model difference between local models and the global model), which can reflect the quality of the model, and calculate the aggregation weight for edge devices based on the key indices. Furthermore, we dynamically adjust aggregation weight based on accuracy's variety to solve weight staleness during the training process. Experiments on the MNIST, FMNIST, EMNIST, CINIC-10 and CIFAR-10 datasets show that the FedDynamic algorithm has better accuracy and convergence performance, compared to the FedAvg, FedProx and Scaffold algorithms.
关键词	Federated Learning Model Aggregation Non-IID
DOI	10.1093/comjnl/bxac118
收录类别	SCI
语种	英语
资助项目	Technology Research and Development Program of China[2019YFB2102100] ; National Natural Science Foundation of China[62072146] ; National Natural Science Foundation of China[61972358] ; Key Research and Development Program of Zhejiang Province[2021C03187] ; Key Research and Development Program of Zhejiang Province[2019C03134]
WOS研究方向	Computer Science
WOS类目	Computer Science, Hardware & Architecture ; Computer Science, Information Systems ; Computer Science, Software Engineering ; Computer Science, Theory & Methods
WOS记录号	WOS:000862225700001
出版者	OXFORD UNIV PRESS
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/19815
专题	中国科学院计算技术研究所期刊论文
通讯作者	Wan, Jian; Ren, Yongjian
作者单位	1.Hangzhou Dianzi Univ, Sch Comp Sci & Technol, Hangzhou 310018, Peoples R China 2.Minist Educ, Key Lab Complex Syst Modeling & Simulat, Hangzhou 310018, Peoples R China 3.Zhejiang Engn Res Ctr Data Secur Governance, Hangzhou 310018, Peoples R China 4.Hangzhou Dianzi Univ, HDU ITMO Joint Inst, Hangzhou 310018, Peoples R China 5.Chinese Acad Sci, Inst Comp Technol, Beijing 100086, Peoples R China
推荐引用方式 GB/T 7714	Zeng, Yan,Mu, Yuankai,Yuan, Junfeng,et al. Adaptive Federated Learning With Non-IID Data[J]. COMPUTER JOURNAL,2022:15.
APA	Zeng, Yan.,Mu, Yuankai.,Yuan, Junfeng.,Teng, Siyuan.,Zhang, Jilin.,...&Zhang, Yunquan.(2022).Adaptive Federated Learning With Non-IID Data.COMPUTER JOURNAL,15.
MLA	Zeng, Yan,et al."Adaptive Federated Learning With Non-IID Data".COMPUTER JOURNAL (2022):15.