Aiding Convergence in Federated Learning via Local Perturbation and Mutual Similarity Information

Federated learning has emerged in the last decade as a distributed optimization paradigm due to the rapidly increasing number of portable devices able to support the heavy computational needs related to the training of machine learning models. Federated learning utilizes gradient-based optimization to minimize a loss objective shared across participating agents. In this paper, we propose a novel modified federated-learning framework wherein each client locally performs a perturbed gradient step leveraging prior information about other statistically affine clients. We theoretically prove that our procedure, due to a suitably introduced adaptation in the update rule, achieves a quantifiable speedup on the exponential contraction factor in the strongly convex case compared with popular algorithms FedAvg and FedProx, here analyzed as baselines. Lastly, we legitimize our conclusions through experimental results on the CIFAR10 and FEMNIST datasets, where we show that our algorithm speeds convergence up to a margin of 30 global rounds compared with FedAvg while modestly improving generalization on unseen data in heterogeneous settings.