Abstract
We propose a federated reinforcement learning (FedRL) solution to innovate resource allocation in converged access networks. FedRL lowers network latency with reinforcement-learnt bandwidth decision and achieves fast learning with federated learning efforts.
© 2020 The Author(s)
PDF ArticleMore Like This
Qi Zhou, Jingjie Zhu, Junwen Zhang, Zhensheng Jia, Bernardo Huberman, and Gee-Kung Chang
STh3L.1 CLEO: Science and Innovations (CLEO:S&I) 2020
Yingmin Zuo, Yongli Zhao, Xiaosong Yu, Avishek Nag, and Jie Zhang
T3C.6 Asia Communications and Photonics Conference (ACP) 2020
Elaine Wong and Lihua Ruan
Mo3C.5 European Conference and Exhibition on Optical Communication (ECOC) 2022