PFL-MoE: Personalized Federated Learning Based on Mixture of Experts

1 code implementation • 31 Dec 2020 • Binbin Guo, Yuan Mei, Danyang Xiao, Weigang Wu, Ye Yin, Hongli Chang

To achieve model personalization while maintaining generalization, in this paper, we propose a new approach, named PFL-MoE, which mixes outputs of the personalized model and global model via the MoE architecture.

Decision Making Personalized Federated Learning