In this work, we consider the distributed stochastic optimization problem of minimizing a non-convex function $f(x) = mathbb{E}_{xi sim mathcal{D}} f(x; xi)$ in an adversarial setting, where the individual functions $f(x; xi)$ can also be potentially non-convex. We assume that at most $alpha$-fraction of a total of $K$ nodes can be Byzantines. We propose a robust stochastic variance-reduced gradient (SVRG) like algorithm for the problem, where the batch gradients are computed at the worker nodes (WNs) and the stochastic gradients are computed at the server node (SN). For the non-convex optimization problem, we show that we need $tilde{O}left( frac{1}{epsilon^{5/3} K^{2/3}} + frac{alpha^{4/3}}{epsilon^{5/3}} right)$ gradient computations on average at each node (SN and WNs) to reach an $epsilon$-stationary point. The proposed algorithm guarantees convergence via the design of a novel Byzantine filtering rule which is independent of the problem dimension. Importantly, we capture the effect of the fraction of Byzantine nodes $alpha$ present in the network on the convergence performance of the algorithm.
Download