Byzantine-Robust Federated Learning with Optimal Statistical Rates

We propose Byzantine-robust federated learning protocols with nearly optimal statistical rates based on recent progress in high dimensional robust statistics. In contrast to prior work, our proposed protocols improve the dimension dependence and achieve a tight statistical rate in terms of all the parameters for strongly convex losses. We also provide matching statistical lower bound for the problem. For experiments, we benchmark against competing protocols and show the empirical superiority of the proposed protocols.