Personalized Multi-tier Federated Learning

TL;DR

This paper introduces PerMFL, a multi-tier personalized federated learning framework that effectively handles data heterogeneity, guarantees convergence, and outperforms existing methods in various tasks.

Contribution

The paper proposes PerMFL, a novel multi-tier architecture for personalized federated learning with theoretical convergence guarantees and superior empirical performance.

Findings

PerMFL achieves linear convergence for strongly convex problems.

PerMFL attains sub-linear convergence for non-convex problems.

Empirical results show PerMFL outperforms state-of-the-art methods.

Abstract

The key challenge of personalized federated learning (PerFL) is to capture the statistical heterogeneity properties of data with inexpensive communications and gain customized performance for participating devices. To address these, we introduced personalized federated learning in multi-tier architecture (PerMFL) to obtain optimized and personalized local models when there are known team structures across devices. We provide theoretical guarantees of PerMFL, which offers linear convergence rates for smooth strongly convex problems and sub-linear convergence rates for smooth non-convex problems. We conduct numerical experiments demonstrating the robust empirical performance of PerMFL, outperforming the state-of-the-art in multiple personalized federated learning tasks.