FedEL: Federated ensemble learning for non-iid data

Xing Wu, Jie Pei, Xian-Hua Han, Yen-Wei Chen, Junfeng Yao, Yang Liu, Quan Qian, Yike Guo

Abstract: Highlights•We propose a federated ensemble learning method to solve the non-IID problem.•Our method can improve model accuracy and reduce storage and inference costs.•Our approach is effective on both natural scenes and medical image datasets.