Google跑不到谱· Issue #30 · NeymarL/ChineseChess-AlphaZero · GitHub

cchess_alphazero.worker.self_play INFO # Exception Shape must be rank 1 but is rank 0 for 'input_batchnorm_4/cond/Reshape_4' (op: 'Reshape') with input shapes: [1,192,1,1], []., 重新加载权重 2018-11-07 10:16:32,709@cchess_alphazero.agent.model

tensorflow 2.0 学习（十二）卷积神经网络（二） CIFAR10数据集与改进VGG13网络+ CoLab - Z_He - 博客园

AlphaZero: How Intuition Demolished Logic, by Carlos E. Perez, Intuition Machine

动手实现并行版AlphaZero · hijkzzz/alpha-zero-gomoku Wiki · GitHub

AlphaZero Plays a Tal Move Chess Has A Bright Future

Google Deepmind's AlphaZero Chess Engine Strangles Stockfish

玩的时候总是卡住很长时间· Issue #46 · NeymarL/ChineseChess-AlphaZero · GitHub

AlphaGo Zero论文解析

使用PyTorch实现简单的AlphaZero的算法（3）：神经网络架构和自学习_alphazero 网络实现-CSDN博客

最强通用棋类AI，AlphaZero强化学习算法解读- 知乎