Research Area:  Machine Learning

Abstract:

Non-IID data present a tough challenge for federated learning. In this paper, we explore a novel idea of facilitating pairwise collaborations between clients with similar data. We propose FedAMP, a new method employing federated attentive message passing to facilitate similar clients to collaborate more. We establish the convergence of FedAMP for both convex and non-convex models, and propose a heuristic method to further improve the performance of FedAMP when clients adopt deep neural networks as personalized models. Our extensive experiments on benchmark data sets demonstrate the superior performance of the proposed methods.

Keywords:  

Author(s) Name:  Yutao Huang, Lingyang Chu, Zirui Zhou, Lanjun Wang, Jiangchuan Liu, Jian Pei, Yong Zhang

Publisher name:  arXiv:2007.03797

DOI:  10.48550/arXiv.2007.03797

Volume Information:  

Paper Link:   https://arxiv.org/abs/2007.03797