結果 : reinforcement learning and artificial neural networks