Learning by imitating the classics: Mitigating class imbalance in federated learning via simulated centralized learning

Guogang Zhu, Xuefeng Liu, Jianwei Niu, Yucheng Wei, Shaojie Tang, Jiayuan Zhang

Abstract: Highlights•A serial training paradigm to improve the model in federated learning.•FedSCL:A parallel strategy to increase the convergence speed of serial training.•Convergence analysis and experimental verification for FedSCL.