DRL-TPE: Learning to Optimize TPE of Optical Interconnects to Accelerate Hitless Reconfiguration of OCS-based DCNs

We leverage deep reinforcement learning (DRL) to optimize the topology engineering of optical circuit switching based data-center networks, such that the hitless reconfiguration to the obtained target physical topology can be finished with the fewest stages. Simulations confirm the benefits of our proposal over the state-of-the-art.