An Aggregation-Free Federated Learning for Tackling Data Heterogeneity

arXiv:2404.18962v1 Announce Type: cross
Abstract: The performance of Federated Learning (FL) hinges on the effectiveness of utilizing knowledge from distributed datasets. Traditional FL methods adopt an aggregate-then-adapt framework, where clients update local models based on a global model aggregated by the server from the previous training round. This process can cause client drift, especially with significant cross-client data heterogeneity, impacting model performance and convergence of the FL algorithm. To address these challenges, we introduce FedAF, a novel aggregation-free FL algorithm. …

abstract adapt aggregation arxiv client cs.cv cs.lg data datasets distributed federated learning framework free global knowledge performance process server training type update