Adapt to Adaptation: Learning Personalization for Cross-Silo Federated Learning

TL;DR

This paper introduces APPLE, a personalized federated learning framework that adaptively balances global and local models to improve performance across non-IID client data, demonstrating state-of-the-art results.

Contribution

The paper proposes a novel adaptive personalization method for cross-silo federated learning that flexibly balances global and local training objectives.

Findings

APPLE achieves state-of-the-art performance on benchmark datasets.

The method effectively handles non-IID data distributions.

Empirical results show improved convergence and generalization.

Abstract

Conventional federated learning (FL) trains one global model for a federation of clients with decentralized data, reducing the privacy risk of centralized training. However, the distribution shift across non-IID datasets, often poses a challenge to this one-model-fits-all solution. Personalized FL aims to mitigate this issue systematically. In this work, we propose APPLE, a personalized cross-silo FL framework that adaptively learns how much each client can benefit from other clients' models. We also introduce a method to flexibly control the focus of training APPLE between global and local objectives. We empirically evaluate our method's convergence and generalization behaviors, and perform extensive experiments on two benchmark datasets and two medical imaging datasets under two non-IID settings. The results show that the proposed personalized FL framework, APPLE, achieves state-of-the-art performance compared to several other personalized FL approaches in the literature. The code is publicly available at https://github.com/ljaiverson/pFL-APPLE.