Personalized Federated Learning with Server-Side Information

TL;DR

This paper introduces FedSIM, a personalized federated learning method that leverages server-side data to enhance model personalization, achieving higher accuracy and faster convergence with reduced computational costs.

Contribution

The paper proposes FedSIM, a novel approach that utilizes server data to improve meta-gradient computation in personalized federated learning, addressing resource constraints.

Findings

FedSIM outperforms existing methods in accuracy across benchmarks.

FedSIM reduces training time by up to 34.2%.

FedSIM is more computationally efficient by calculating full meta-gradients on the server.

Abstract

Personalized Federated Learning (FL) is an emerging research field in FL that learns an easily adaptable global model in the presence of data heterogeneity among clients. However, one of the main challenges for personalized FL is the heavy reliance on clients' computing resources to calculate higher-order gradients since client data is segregated from the server to ensure privacy. To resolve this, we focus on a problem setting where the server may possess its own data independent of clients' data -- a prevalent problem setting in various applications, yet relatively unexplored in existing literature. Specifically, we propose FedSIM, a new method for personalized FL that actively utilizes such server data to improve meta-gradient calculation in the server for increased personalization performance. Experimentally, we demonstrate through various benchmarks and ablations that FedSIM is superior to existing methods in terms of accuracy, more computationally efficient by calculating the full meta-gradients in the server, and converges up to 34.2% faster.