FedPop: A Bayesian Approach for Personalised Federated Learning
This work addresses uncertainty and personalization challenges in federated learning, particularly for new or small-data clients, offering a novel Bayesian approach with theoretical guarantees.
The paper tackles the lack of uncertainty quantification and issues with new or data-scarce clients in personalized federated learning by proposing FedPop, a Bayesian method that models clients with common parameters and random effects, resulting in robustness to client drift, practical inference for new clients, and enabled uncertainty quantification with mild overhead.
Personalised federated learning (FL) aims at collaboratively learning a machine learning model taylored for each client. Albeit promising advances have been made in this direction, most of existing approaches works do not allow for uncertainty quantification which is crucial in many applications. In addition, personalisation in the cross-device setting still involves important issues, especially for new clients or those having small number of observations. This paper aims at filling these gaps. To this end, we propose a novel methodology coined FedPop by recasting personalised FL into the population modeling paradigm where clients' models involve fixed common population parameters and random effects, aiming at explaining data heterogeneity. To derive convergence guarantees for our scheme, we introduce a new class of federated stochastic optimisation algorithms which relies on Markov chain Monte Carlo methods. Compared to existing personalised FL methods, the proposed methodology has important benefits: it is robust to client drift, practical for inference on new clients, and above all, enables uncertainty quantification under mild computational and memory overheads. We provide non-asymptotic convergence guarantees for the proposed algorithms and illustrate their performances on various personalised federated learning tasks.