SHEFL: Sparse Heterogeneous Ensemble Federated Learning with Group-Balanced Aggregation
Across four vision datasets, SHEFL improves over Fed-Ensemble (FedEns) in all data heterogeneity settings and under a matched total uplink budget, SHEFL reaches the target accuracy in fewer rounds than FedEns and improves final accuracy by up to 14.65 points.
Keumseo Ryum, J. Gong, Joonhyuk Kang
· 0 citations