Learning a locomotion controller for a quadrupedal robot with Deep Reinforcement Learning Presentation Paper