Federated Learning with Superquantile Aggregation for Heterogeneous Data

TL;DR

This paper introduces a federated learning method using superquantile-based objectives to improve robustness across heterogeneous client data, with proven convergence and superior tail error performance.

Contribution

It proposes a novel superquantile aggregation approach for federated learning that enhances robustness and provides convergence guarantees.

Findings

Competitive average error performance

Outperforms in tail error statistics

Proven finite-time convergence guarantees

Abstract

We present a federated learning framework that is designed to robustly deliver good predictive performance across individual clients with heterogeneous data. The proposed approach hinges upon a superquantile-based learning objective that captures the tail statistics of the error distribution over heterogeneous clients. We present a stochastic training algorithm that interleaves differentially private client filtering with federated averaging steps. We prove finite time convergence guarantees for the algorithm: $O(1/\sqrt{T})$ in the nonconvex case in $T$ communication rounds and $O(\exp(-T/\kappa^{3/2}) + \kappa/T)$ in the strongly convex case with local condition number $\kappa$. Experimental results on benchmark datasets for federated learning demonstrate that our approach is competitive with classical ones in terms of average error and outperforms them in terms of tail statistics of the error.