RNN Deep Reinforcement Learning for Routing Optimization
Penghao Sun, Junfei Li, Julong Lan, Yuxiang Hu, Xin Lu · 2018
Routing optimization has been discussed in network design for a long time. In recent years, new methods of routing strategy based on Reinforcement Learning are being considered. In this paper, we propose a reinforcement learning based smart agent that can optimize routing strategy without human experience. Our proposed scheme is based on the collection of the traffic intensity in switches and the usage of a Recurrent Neural Network based deep reinforcement learning model to train the agent. Simulation result shows that the proposed scheme can adjust the routing strategy dynamically according to the network condition and outperforms the traditional shortest path routing after trained.