Adapt to Adaptation: Learning Personalization for Cross-Silo Federated Learning
This work addresses the problem of non-IID data in federated learning for clients like medical institutions, offering an incremental improvement over existing personalized FL methods.
The paper tackles the challenge of distribution shift in federated learning by proposing APPLE, a personalized cross-silo framework that adaptively learns client benefits and flexibly controls training focus, achieving state-of-the-art performance on benchmark and medical imaging datasets under non-IID settings.
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.