Alphago是用reinforcement learning,估计几张4090就够了

请您先登陆,再发跟帖!