FedSelect: Customized Selection of Parameters for Fine-Tuning during Personalized Federated Learning

TL;DR

FedSelect is a novel federated learning framework that dynamically identifies optimal client subnetworks for personalized fine-tuning, balancing local adaptation with global knowledge retention, leading to improved performance on CIFAR-10.

Contribution

The paper introduces a new method for personalized federated learning that discovers optimal client subnetworks for fine-tuning, inspired by the lottery ticket hypothesis, enhancing personalization without sacrificing global knowledge.

Findings

Achieves promising results on CIFAR-10.

Balances local personalization with global knowledge retention.

Introduces a subnetwork discovery approach for FL.

Abstract

Recent advancements in federated learning (FL) seek to increase client-level performance by fine-tuning client parameters on local data or personalizing architectures for the local task. Existing methods for such personalization either prune a global model or fine-tune a global model on a local client distribution. However, these existing methods either personalize at the expense of retaining important global knowledge, or predetermine network layers for fine-tuning, resulting in suboptimal storage of global knowledge within client models. Enlightened by the lottery ticket hypothesis, we first introduce a hypothesis for finding optimal client subnetworks to locally fine-tune while leaving the rest of the parameters frozen. We then propose a novel FL framework, FedSelect, using this procedure that directly personalizes both client subnetwork structure and parameters, via the simultaneous discovery of optimal parameters for personalization and the rest of parameters for global aggregation during training. We show that this method achieves promising results on CIFAR-10.