Adaptive Personalized Federated Learning

TL;DR

This paper introduces an adaptive personalized federated learning algorithm that balances local and global model training, with theoretical analysis and experiments showing improved personalization and communication efficiency.

Contribution

It proposes an adaptive federated learning method with a derived generalization bound, optimal mixing parameter, and convergence analysis in various settings.

Findings

Effective personalization demonstrated through experiments

Optimal mixing parameter derived for model combination

Communication-efficient optimization method validated

Abstract

Investigation of the degree of personalization in federated learning algorithms has shown that only maximizing the performance of the global model will confine the capacity of the local models to personalize. In this paper, we advocate an adaptive personalized federated learning (APFL) algorithm, where each client will train their local models while contributing to the global model. We derive the generalization bound of mixture of local and global models, and find the optimal mixing parameter. We also propose a communication-efficient optimization method to collaboratively learn the personalized models and analyze its convergence in both smooth strongly convex and nonconvex settings. The extensive experiments demonstrate the effectiveness of our personalization schema, as well as the correctness of established generalization theories.