FedTSA: A Cluster-based Two-Stage Aggregation Method for Model-heterogeneous Federated Learning

TL;DR

FedTSA introduces a novel cluster-based two-stage aggregation method for federated learning that effectively handles system heterogeneity by grouping clients and employing different aggregation strategies for homogeneous and heterogeneous models.

Contribution

This paper proposes FedTSA, a new method that clusters clients by capabilities and uses a two-stage aggregation to improve model performance in resource-diverse federated learning environments.

Findings

FedTSA outperforms baseline methods in heterogeneous settings.

Clustering clients improves aggregation efficiency and model accuracy.

Two-stage aggregation effectively handles model heterogeneity.

Abstract

Despite extensive research into data heterogeneity in federated learning (FL), system heterogeneity remains a significant yet often overlooked challenge. Traditional FL approaches typically assume homogeneous hardware resources across FL clients, implying that clients can train a global model within a comparable time frame. However, in practical FL systems, clients often have heterogeneous resources, which impacts their training capacity. This discrepancy underscores the importance of exploring model-heterogeneous FL, a paradigm allowing clients to train different models based on their resource capabilities. To address this challenge, we introduce FedTSA, a cluster-based two-stage aggregation method tailored for system heterogeneity in FL. FedTSA begins by clustering clients based on their capabilities, then performs a two-stage aggregation: conventional weight averaging for homogeneous models in Stage 1, and deep mutual learning with a diffusion model for aggregating heterogeneous models in Stage 2. Extensive experiments demonstrate that FedTSA not only outperforms the baselines but also explores various factors influencing model performance, validating FedTSA as a promising approach for model-heterogeneous FL.