Adaptive asynchronous federated learning

Renhao Lu, Weizhe Zhang, Qiong Li, Hui He, Xiaoxiong Zhong, Hongwei Yang, Desheng Wang, Zenglin Xu, Mamoun Alazab

Abstract: Highlights•Design a novel adaptive asynchronous federated learning framework with momentum.•Propose an adaptive weight allocation algorithm for the asynchronous model update.•Investigate the impact of dynamic environments on federated learning.•Prove the convergence rate of the proposed asynchronous federated learning scheme.•The proposed FL scheme can cope with the system load’s dynamics.