Byzantine-Robust Decentralized Learning via ClippedGossip

In this paper, we study the challenging task of Byzantine-robust decentralized training on arbitrary communication graphs. Unlike federated learning where workers communicate through a server, workers in the decentralized environment can only talk to their neighbors, making it harder to reach consensus and benefit from collaborative training. To address these issues, we propose a ClippedGossip algorithm for Byzantine-robust consensus and optimization, which is the first to provably converge to a $O({\delta }_{\max }{\zeta }^2\mathrm{/}{\gamma }^2)$ neighborhood of the stationary point for non-convex objectives under standard assumptions. Finally, we demonstrate the encouraging empirical performance of ClippedGossip under a large number of attacks.