D Deep Reinforcement Learning (7)