Reinforcement Learning for Stochastic Networks - Sciencesconf.org