Controlling Participation in Federated Learning with Feedback

TL;DR

This paper introduces FedBack, a control-theoretic method for managing client participation in federated learning, leading to significant efficiency improvements over random client selection.

Contribution

FedBack is a novel deterministic approach using control theory to optimize client participation in federated learning with proven convergence guarantees.

Findings

Up to 50% improvement in communication efficiency

Effective control of client participation dynamics

Global convergence guarantees established

Abstract

We address the problem of client participation in federated learning, where traditional methods typically rely on a random selection of a small subset of clients for each training round. In contrast, we propose FedBack, a deterministic approach that leverages control-theoretic principles to manage client participation in ADMM-based federated learning. FedBack models client participation as a discrete-time dynamical system and employs an integral feedback controller to adjust each client's participation rate individually, based on the client's optimization dynamics. We provide global convergence guarantees for our approach by building on the recent federated learning research. Numerical experiments on federated image classification demonstrate that FedBack achieves up to 50\% improvement in communication and computational efficiency over algorithms that rely on a random selection of clients.