C^(2)S:Class-aware client selection for effective aggregation in federated learning

作     者：Mei Cao Yujie Zhang Zezhong Ma Mengying Zhao 

作者机构：School of Computer Science and TechnologyShandong UniversityJinanChina State Key Laboratory of High-End Server&Storage TechnologyJinanChina 

出 版 物：《High-Confidence Computing》 (高置信计算（英文）)

年 卷 期：2022年第2卷第3期

页      面：7-15页

核心收录：

学科分类：08[工学] 081201[工学-计算机系统结构] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by Shandong Provincial Natural Sci-ence Foundation,China(ZR2020LZH001) NSFC-Shandong Joint Fund,China(U1806203) Major scientific and technological innovation project in Shandong Province,China(2019JZZY010449) 

主　　题：Federated learning Non-IID data Aggregation Client selection 

摘      要：Federated learning is proposed to train distributed data in a safe manner by avoiding to send data to *** server maintains a global model and sends it to clients in each communication round,and then aggregates the updated local models to derive a new global ***,the clients are randomly selected in each round and aggregation is based on weighted *** show that the performance on IID data is satisfactory while significant accuracy drop can be observed for Non-IID *** this paper,we explore the reasons and propose a novel aggregation approach for Non-IID data in federated ***,we propose to group the clients according to classes of data they have,and select one set in each communication *** models from the same set are averaged as usual and the updated global model is sent to next group of clients for further *** this way,the parameters are only averaged on similar clients and passed among different *** shows that the proposed scheme has advantages in terms of model accuracy and convergence speed with highly unbalanced data distribution and complex models.