Routing in quantum communication networks using reinforcement machine learning

This paper promotes reinforcement machine learning for route-finding tasks in quantum communication networks, where, due to the non-additivity of quantum errors, classical graph path or tree-finding algorithms cannot be used. We propose using a proximal policy optimization algorithm capable of finding routes in teleportation-based quantum networks. This algorithm is benchmarked against the Monte Carlo search. The topology of our network resembles the proposed 6 G topology and analyzed that quantum errors correspond to typical errors in realistic quantum channels.