CASIA OpenIR  > 多模态人工智能系统全国重点实验室
CFSL: A Credible Federated Self-Learning Framework
Zhang, Weishan1; Bao, Zhicheng1; Liu, Yuru1; Xu, Liang2; Lu, Qinghua3; Ning, Huansheng2; Wang, Xiao4; Yang, Su5; Wang, Fei-Yue4; Li, Zengxiang6
Source PublicationIEEE INTERNET OF THINGS JOURNAL
ISSN2327-4662
2023-12-15
Volume10Issue:24Pages:21349-21362
Corresponding AuthorZhang, Weishan(zhangws@upc.edu.cn)
AbstractFederated learning can collaboratively train AI models while protecting data privacy. In practical industry environment, non-independent and identically distributed (Non-IID) characteristics of data affect the effectiveness of federated learning. Personalized federated learning can help resolve this, but it cannot adapt to unknown data. In addition, practical applications also call for trusted training environment and remain stable when there are security threats. In this article, we propose a credible federated self-learning (CFSL), based on the idea of hypernetwork supported by blockchain to achieve secured, credible, personalized federated self-learning, especially, for unknown data in Non-IID environment. Extensive experiments on three Non-IID data sets demonstrate the capabilities on adaptive resilience for security attacks and on accuracy of recognizing unknown objects, with good performance at the same time. CFSL outperforms the existing personalized federated learning methods, with an increase in average accuracy by 4.11%.
KeywordFederated learning Data models Training Internet of Things Blockchains Adaptation models Smart contracts Blockchain consensus federated learning personalization self-learning
DOI10.1109/JIOT.2023.3286398
WOS KeywordINTERNET
Indexed BySCI
Language英语
Funding ProjectNational Natural Science Foundation of China
Funding OrganizationNational Natural Science Foundation of China
WOS Research AreaComputer Science ; Engineering ; Telecommunications
WOS SubjectComputer Science, Information Systems ; Engineering, Electrical & Electronic ; Telecommunications
WOS IDWOS:001142524100004
PublisherIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Citation statistics
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/55441
Collection多模态人工智能系统全国重点实验室
Corresponding AuthorZhang, Weishan
Affiliation1.China Univ Petr East China, Coll Comp Sci & Technol, Qingdao 266580, Peoples R China
2.Beijing Univ Sci & Technol, Coll Comp & Commun Engn, Beijing 100083, Peoples R China
3.CSIRO, Data61, Sydney, NSW 2070, Australia
4.Chinese Acad Sci, Inst Automat, Beijing 100080, Peoples R China
5.Fudan Univ, Coll Comp Sci, Shanghai 200437, Peoples R China
6.ENN Grp, Inst Digital Res, Langfang 065001, Peoples R China
Recommended Citation
GB/T 7714
Zhang, Weishan,Bao, Zhicheng,Liu, Yuru,et al. CFSL: A Credible Federated Self-Learning Framework[J]. IEEE INTERNET OF THINGS JOURNAL,2023,10(24):21349-21362.
APA Zhang, Weishan.,Bao, Zhicheng.,Liu, Yuru.,Xu, Liang.,Lu, Qinghua.,...&Li, Zengxiang.(2023).CFSL: A Credible Federated Self-Learning Framework.IEEE INTERNET OF THINGS JOURNAL,10(24),21349-21362.
MLA Zhang, Weishan,et al."CFSL: A Credible Federated Self-Learning Framework".IEEE INTERNET OF THINGS JOURNAL 10.24(2023):21349-21362.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhang, Weishan]'s Articles
[Bao, Zhicheng]'s Articles
[Liu, Yuru]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhang, Weishan]'s Articles
[Bao, Zhicheng]'s Articles
[Liu, Yuru]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhang, Weishan]'s Articles
[Bao, Zhicheng]'s Articles
[Liu, Yuru]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.