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

The goal of conventional federated learning (FL) is to train a global model for a federation of clients with decentralized data, reducing the systemic privacy risk of centralized training. The distribution shift across non-IID datasets, also known as the data heterogeneity, often poses a challenge for this one-global-model-fits-all solution. 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 behavior and performed 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.