A Systematic Literature Review of Blockchain-based Federated Learning: Architectures, Applications and Issues
Dongkun Hou, Jie Zhang, Ka Lok Man, Jieming Ma, Zitian Peng
Abstract
Federal learning (FL) can realize a distributed training machine learning models in multiple devices while protecting their data privacy, but some defect still exists such as single point failure and lack of motivation. Blockchain as a distributed ledger can be utilized to provide a novel FL framework to address those issues. This paper aims to discuss how the blockchain technology is employed to compensate for shortcomings in FL. A systematic literature review is conducted to investigate existing FL problems and to summarize knowledge about the existing Blockchain-based FL (BFL). The differences among these collected BFL architectures are presented and discussed, and the applications of BFL are categorized and analyzed. Finally, some suggestions for future development and application of BFL are discussed.