Variance-Reduced Heterogeneous Federated Learning via Stratified Client Selection

TL;DR

This paper introduces a stratified client selection method for federated learning that reduces variance caused by data heterogeneity, leading to improved convergence and accuracy.

Contribution

It proposes a novel stratified selection scheme based on client data distribution and an optimized sample size allocation to enhance federated learning performance.

Findings

Achieves lower variance compared to existing methods.

Improves convergence speed and accuracy.

Compatible with common federated learning algorithms.

Abstract

Client selection strategies are widely adopted to handle the communication-efficient problem in recent studies of Federated Learning (FL). However, due to the large variance of the selected subset's update, prior selection approaches with a limited sampling ratio cannot perform well on convergence and accuracy in heterogeneous FL. To address this problem, in this paper, we propose a novel stratified client selection scheme to reduce the variance for the pursuit of better convergence and higher accuracy. Specifically, to mitigate the impact of heterogeneity, we develop stratification based on clients' local data distribution to derive approximate homogeneous strata for better selection in each stratum. Concentrating on a limited sampling ratio scenario, we next present an optimized sample size allocation scheme by considering the diversity of stratum's variability, with the promise of further variance reduction. Theoretically, we elaborate the explicit relation among different selection schemes with regard to variance, under heterogeneous settings, we demonstrate the effectiveness of our selection scheme. Experimental results confirm that our approach not only allows for better performance relative to state-of-the-art methods but also is compatible with prevalent FL algorithms.