典型文献
A Survey of Federated Learning on Non-IID Data
文献摘要：
Federated learning (FL) is a machine learning paradigm for data silos and privacy protection,which aims to organize multiple cli-ents for training global machine learning models without exposing data to all parties. However, when dealing with non-independently identi-cally distributed (non-IID) client data, FL cannot obtain more satisfactory results than centrally trained machine learning and even fails to match the accuracy of the local model obtained by client training alone. To analyze and address the above issues, we survey the state-of-the-art methods in the literature related to FL on non-IID data. On this basis, a motivation-based taxonomy, which classifies these methods into two categories, including heterogeneity reducing strategies and adaptability enhancing strategies, is proposed. Moreover, the core ideas and main challenges of these methods are analyzed. Finally, we envision several promising research directions that have not been thoroughly stud-ied, in hope of promoting research in related fields to a certain extent.
文献关键词：
中图分类号：
作者姓名：
HAN Xuming;GAO Minghan;WANG Limin;HE Zaobo;WANG Yanze
作者机构：
Jinan University, Guangzhou 510632, China;Changchun University of Technology, Changchun 130012, China;Guangdong University of Finance&Economics, Guangzhou 510320, China
文献出处：
引用格式：
[1]HAN Xuming;GAO Minghan;WANG Limin;HE Zaobo;WANG Yanze-.A Survey of Federated Learning on Non-IID Data)[J].中兴通讯技术（英文版）,2022(03):17-26
A类：
centrally
B类：
Survey,Federated,Learning,Non,IID,Data,learning,FL,machine,paradigm,data,silos,privacy,protection,which,aims,organize,multiple,ents,training,global,models,without,exposing,parties,However,when,dealing,independently,identi,cally,distributed,client,cannot,more,satisfactory,results,than,trained,even,fails,match,accuracy,local,obtained,by,alone,To,address,above,issues,survey,state,methods,literature,related,On,this,basis,motivation,taxonomy,classifies,these,into,two,categories,including,heterogeneity,reducing,strategies,adaptability,enhancing,proposed,Moreover,core,ideas,main,challenges,are,analyzed,Finally,envision,several,promising,research,directions,that,have,been,thoroughly,stud,ied,hope,promoting,fields,certain,extent
AB值：
0.670632
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
