Controlling Participation in Federated Learning with Feedback
This addresses efficiency issues in federated learning for distributed systems, though it is incremental as it builds on existing ADMM-based methods.
The paper tackles the problem of client participation in federated learning by proposing FedBack, a deterministic control-theoretic approach that adjusts each client's participation rate individually, resulting in up to 50% improvement in communication and computational efficiency over random selection methods.
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.