Provably Personalized and Robust Federated Learning

Abstract

Clustering clients with similar objectives and learning a model per cluster is an intuitive and interpretable approach to personalization in federated learning. However, doing so with provable and optimal guarantees has remained an open challenge. In this work, we formalize personalized federated learning as a stochastic optimization problem. We propose simple clustering-based algorithms which iteratively identify and train within clusters, using local client gradients. Our algorithms have optimal convergence rates which asymptotically match those obtained if we knew the true underlying clustering of the clients, and are provably robust in the Byzantine setting where some fraction of the clients are malicious.