FedStale: leveraging stale client updates in federated learning

TL;DR

FedStale introduces a new federated learning algorithm that optimally combines fresh and stale client updates, improving convergence and performance in heterogeneous participation scenarios.

Contribution

The paper proposes FedStale, a novel algorithm that adaptively combines stale and fresh updates, extending existing methods to handle heterogeneous client participation effectively.

Findings

FedStale outperforms FedAvg and FedVARP in various heterogeneity settings.

The usefulness of stale updates decreases with less data heterogeneity and more participation heterogeneity.

The analysis reveals how the least participating client impacts convergence error.

Abstract

Federated learning algorithms, such as FedAvg, are negatively affected by data heterogeneity and partial client participation. To mitigate the latter problem, global variance reduction methods, like FedVARP, leverage stale model updates for non-participating clients. These methods are effective under homogeneous client participation. Yet, this paper shows that, when some clients participate much less than others, aggregating updates with different levels of staleness can detrimentally affect the training process. Motivated by this observation, we introduce FedStale, a novel algorithm that updates the global model in each round through a convex combination of "fresh" updates from participating clients and "stale" updates from non-participating ones. By adjusting the weight in the convex combination, FedStale interpolates between FedAvg, which only uses fresh updates, and FedVARP, which treats fresh and stale updates equally. Our analysis of FedStale convergence yields the following novel findings: i) it integrates and extends previous FedAvg and FedVARP analyses to heterogeneous client participation; ii) it underscores how the least participating client influences convergence error; iii) it provides practical guidelines to best exploit stale updates, showing that their usefulness diminishes as data heterogeneity decreases and participation heterogeneity increases. Extensive experiments featuring diverse levels of client data and participation heterogeneity not only confirm these findings but also show that FedStale outperforms both FedAvg and FedVARP in many settings.