Accelerating Byzantine-Robust Distributed Learning with Compressed Communication via Double Momentum and Variance Reduction

AISTATS '26