Data_Sheet_1_Learning to Play the Chess Variant Crazyhouse Above World Champion Level With Deep Neural Networks and Human Data.ZIP

Deep neural networks have been successfully applied in learning the board games Go, chess, and shogi without prior knowledge by making use of reinforcement learning. Although starting from zero knowledge has been shown to yield impressive results, it is associated with high computationally costs especially for complex games. With this paper, we present CrazyAra which is a neural network based engine solely trained in supervised manner for the chess variant crazyhouse. Crazyhouse is a game with a higher branching factor than chess and there is only limited data of lower quality available compared to AlphaGo. Therefore, we focus on improving efficiency in multiple aspects while relying on low computational resources. These improvements include modifications in the neural network design and training configuration, the introduction of a data normalization step and a more sample efficient Monte-Carlo tree search which has a lower chance to blunder. After training on 569537 human games for 1.5 days we achieve a move prediction accuracy of 60.4%. During development, versions of CrazyAra played professional human players. Most notably, CrazyAra achieved a four to one win over 2017 crazyhouse world champion Justin Tan (aka LM Jann Lee) who is more than 400 Elo higher rated compared to the average player in our training set. Furthermore, we test the playing strength of CrazyAra on CPU against all participants of the second Crazyhouse Computer Championships 2017, winning against twelve of the thirteen participants. Finally, for CrazyAraFish we continue training our model on generated engine games. In 10 long-time control matches playing Stockfish 10, CrazyAraFish wins three games and draws one out of 10 matches.

深度神经网络已被成功应用于学习围棋、象棋和将棋等棋类游戏，且无需先验知识，通过强化学习的方式进行。尽管从零知识出发已被证明能够取得令人瞩目的成果，但这与高昂的计算成本相关，尤其是在复杂游戏中。在本论文中，我们提出了CrazyAra，这是一个基于神经网络的引擎，专门以监督学习的方式进行训练，用于棋类变体crazyhouse。与象棋相比，crazyhouse具有更高的分支因子，且相较于AlphaGo，可用的低质量数据有限。因此，我们在依赖较低的计算资源的同时，着重于从多个方面提升效率。这些改进包括对神经网络设计和训练配置的调整、引入数据归一化步骤以及更高效的蒙特卡洛树搜索，后者犯错的概率更低。经过在569537场人类对弈数据上1.5天的训练，我们实现了60.4%的走棋预测准确率。在开发过程中，CrazyAra的版本曾与专业人类选手进行过对弈。最引人注目的是，CrazyAra在与2017年crazyhouse世界冠军Justin Tan（别名LM Jann Lee）的对弈中取得了4比1的胜利，而Justin Tan的Elo等级分比我们训练集中平均选手高出400多分。此外，我们还测试了CrazyAra在CPU上的对弈实力，与2017年第二场Crazyhouse计算机锦标赛的所有参赛者进行了对决，在13位参赛者中击败了12位。最后，对于CrazyAraFish，我们继续在生成的引擎对弈中训练我们的模型。在10场长时间控制对弈中，与Stockfish 10进行对弈，CrazyAraFish赢得了3场比赛，并在10场比赛中平了1场。