Abstract
We design and evaluate DeepAutonet, a self-driving reconfigurable network exploiting deep reinforcement learning. Simulation results show that DeepAutonet can adapt its topology automatically to different traffic conditions with up to 1.9x latency reduction.
© 2019 The Author(s)
PDF ArticleMore Like This
Wen Yang, Bingli Guo, Yu Shang, and Shanguo Huang
M4A.196 Asia Communications and Photonics Conference (ACP) 2020
Baojia Li, Wei Lu, and Zuqing Zhu
Th1H.2 Optical Fiber Communication Conference (OFC) 2019
José Suárez-Varela, Albert Mestres, Junlin Yu, Li Kuang, Haoyu Feng, Pere Barlet-Ros, and Albert Cabellos-Aparicio
M2A.6 Optical Fiber Communication Conference (OFC) 2019