HeterPS: Distributed deep learning with reinforcement learning based scheduling in heterogeneous environments

Ji Liu, Zhihua Wu, Danlei Feng, Minxu Zhang, Xinxuan Wu, Xuefeng Yao, Dianhai Yu, Yanjun Ma, Feng Zhao, Dejing Dou

Abstract: Highlights•The design of HeterPS for the distributed training with heterogeneous resources.•A novel Reinforcement Learning (RL)-based layer scheduling method.•A dynamic method to determine the proper number of computing resources.•An extensive evaluation based on multiple DNN models with divers baseline approaches.