论文标题
使用深入的增强学习来建立三人Mahjong AI
Building a 3-Player Mahjong AI using Deep Reinforcement Learning
论文作者
论文摘要
Mahjong是19世纪后期在中国开发的一款流行的多游戏不完美信息游戏，具有一些非常具有挑战性的AI研究功能。 Sanma是日本Riichi Mahjong的三人游戏变体，具有独特的特征，包括较少的瓷砖，因此具有更具侵略性的游戏风格。因此，它本身就是具有挑战性的，并且具有巨大的研究兴趣，但尚未探索。在本文中，我们介绍了Meowjong，这是Sanma的AI，使用深入的增强学习。我们定义了一个信息丰富且紧凑的二维数据结构，用于编码Sanma游戏中可观察到的信息。我们将5个卷积神经网络（CNN）预先用于SANMA的5个动作 - 丢弃，PON，KAN，KITA和RIICHI，并通过使用Monte Carlo Politight梯度方法来增强主要动作模型，即丢弃模型。 Meowjong的模型实现了通过监督学习与AIS相当的4名玩家Mahjong的测试精度，并从增强学习中获得了重大进一步的增强。作为桑玛（Sanma）有史以来的第一个AI，我们声称Meowjong是这场比赛中最先进的。
Mahjong is a popular multi-player imperfect-information game developed in China in the late 19th-century, with some very challenging features for AI research. Sanma, being a 3-player variant of the Japanese Riichi Mahjong, possesses unique characteristics including fewer tiles and, consequently, a more aggressive playing style. It is thus challenging and of great research interest in its own right, but has not yet been explored. In this paper, we present Meowjong, an AI for Sanma using deep reinforcement learning. We define an informative and compact 2-dimensional data structure for encoding the observable information in a Sanma game. We pre-train 5 convolutional neural networks (CNNs) for Sanma's 5 actions -- discard, Pon, Kan, Kita and Riichi, and enhance the major action's model, namely the discard model, via self-play reinforcement learning using the Monte Carlo policy gradient method. Meowjong's models achieve test accuracies comparable with AIs for 4-player Mahjong through supervised learning, and gain a significant further enhancement from reinforcement learning. Being the first ever AI in Sanma, we claim that Meowjong stands as a state-of-the-art in this game.
