On the Decentralization of Blockchain-enabled Asynchronous Federated Learning

Federated learning (FL), thanks in part to the emergence of the edge computing paradigm, is expected to enable collaborative learning-based applications. However, its original dependence on a central server for orchestration raises several concerns in terms of security, privacy, and scalability. To solve some of these worries, blockchain technology is expected to bring decentralization, robustness, and enhanced trust to FL. The empowerment of FL through blockchain (widely known as FLchain), however, has some implications in terms of ledger inconsistencies that lead to forks and staleness, which are naturally inherited from the blockchain’s fully decentralized operation. Such issues stem from the fact that, given the temporary ledger versions in the blockchain, FL devices may use different models for training, and that, given the asynchronicity of the FL operation, stale local updates (computed using outdated models) may be generated. In this paper, we shed light on the implications of the FLchain setting and study how decentralization in blockchain affects the age of information (AoI) and FL accuracy. To that end, we provide a faithful simulation tool that allows capturing the decentralized and asynchronous nature of the FLchain operation.