Deep Learning in a Nutshell: Reinforcement Learning | Parallel Forall