Blockchain-Enabled Federated Learning: A Reference Architecture Design, Implementation, and Verification

This paper presents an innovative reference architecture for blockchain-enabled federated learning (BCFL), a state-of-the-art approach that amalgamates the strengths of federated learning and blockchain technology. This results in a decentralized, collaborative machine learning system that respects data privacy and user-controlled identity. Our architecture strategically employs a decentralized identifier (DID)-based authentication system, allowing participants to authenticate and then gain access to the federated learning platform securely using their self-sovereign DIDs, which are recorded on the blockchain. Ensuring robust security and efficient decentralization through the execution of smart contracts is a key aspect of our approach. Moreover, our BCFL reference architecture provides significant extensibility, accommodating the integration of various additional elements, as per specific requirements and use cases, thereby rendering it an adaptable solution for a wide range of BCFL applications. Participants can authenticate and then gain access to the federated learning platform securely using their self-sovereign DIDs, which are securely recorded on the blockchain. The pivotal contribution of this study is the successful implementation and validation of a realistic BCFL reference architecture, marking a significant milestone in the field. We intend to make the source code publicly accessible shortly, fostering further advancements and adaptations within the community. This research not only bridges a crucial gap in the current literature but also lays a solid foundation for future explorations in the realm of BCFL.