Deep Deterministic Policy Gradient (DDPG) for learning the optimal policy in continuous action spaces. This is my second project as part of the Deep Reinforcement Learning Nanodegree by Udacity. There are 20 agents in the environments and