Mobility-Aware Asynchronous Federated Learning with Dynamic Sparsification

TL;DR

This paper introduces a mobility-aware dynamic sparsification algorithm for asynchronous federated learning that adapts to device contact patterns and staleness, improving convergence and accuracy in mobile environments.

Contribution

It develops a theoretical model linking sparsification, staleness, and mobility, and proposes a novel MADS algorithm with closed-form solutions for adaptive sparsification.

Findings

MADS improves CIFAR-10 classification accuracy by 8.76%.

MADS reduces trajectory prediction error by 9.46%.

Theoretical analysis guides optimal sparsification based on mobility conditions.

Abstract

Asynchronous Federated Learning (AFL) enables distributed model training across multiple mobile devices, allowing each device to independently update its local model without waiting for others. However, device mobility introduces intermittent connectivity, which necessitates gradient sparsification and leads to model staleness, jointly affecting AFL convergence. This paper develops a theoretical model to characterize the interplay among sparsification, model staleness and mobility-induced contact patterns, and their joint impact on AFL convergence. Based on the analysis, we propose a mobility-aware dynamic sparsification (MADS) algorithm that optimizes the sparsification degree based on contact time and model staleness. Closed-form solutions are derived, showing that under low-speed conditions, MADS increases the sparsification degree to enhance convergence, while under high-speed conditions, it reduces the sparsification degree to guarantee reliable uploads within limited contact time. Experimental results validate the theoretical findings. Compared with the state-of-the-art benchmarks, the MADS algorithm increases the image classification accuracy on the CIFAR-10 dataset by 8.76% and reduces the average displacement error in the Argoverse trajectory prediction dataset by 9.46%.