A study of a federated learning framework based on the interstellar file system and blockchain: Private Blockchain Federated Learning

Federated learning is a scalable machine learning paradigm in which a large group of individuals collaborates to build a high-quality machine learning model. This paper proposes Private Blockchain Federated Learning (PriChainFL). It is a federated learning method that combines blockchain and IPFS, in response to the fact that federated learning is still vulnerable to attacks such as data poisoning and member inference attacks. It also reduces the storage cost of blockchain.