HeteroFL: Computation and Communication Efficient Federated Learning for Heterogeneous Clients

TL;DR

HeteroFL introduces a federated learning framework that efficiently trains heterogeneous local models with varying complexities across clients, producing a unified global model while reducing computation and communication costs.

Contribution

This work presents the first federated learning approach allowing clients with different model architectures and capabilities to collaboratively train a single global model.

Findings

Adaptive subnetwork distribution improves efficiency

Method supports diverse client capabilities

Extensive experiments validate effectiveness

Abstract

Federated Learning (FL) is a method of training machine learning models on private data distributed over a large number of possibly heterogeneous clients such as mobile phones and IoT devices. In this work, we propose a new federated learning framework named HeteroFL to address heterogeneous clients equipped with very different computation and communication capabilities. Our solution can enable the training of heterogeneous local models with varying computation complexities and still produce a single global inference model. For the first time, our method challenges the underlying assumption of existing work that local models have to share the same architecture as the global model. We demonstrate several strategies to enhance FL training and conduct extensive empirical evaluations, including five computation complexity levels of three model architecture on three datasets. We show that adaptively distributing subnetworks according to clients' capabilities is both computation and communication efficient.