Using CNN for solving two-player zero-sum games

作者：

Highlights：

• We use a novel CNN method to solve two-player zero-sum games.

• Concrete training algorithms are proposed to train the CNN model for the games.

• Our CNN model can handle different game sizes and untrained generation distributions.

• Our CNN approach shows great potential in terms of computational efficiency.

摘要

•We use a novel CNN method to solve two-player zero-sum games.•Concrete training algorithms are proposed to train the CNN model for the games.•Our CNN model can handle different game sizes and untrained generation distributions.•Our CNN approach shows great potential in terms of computational efficiency.

论文关键词：Two-player zero-sum game,Saddle point,Convolutional neural network,Machine learning

论文评审过程：Received 20 April 2021, Revised 23 March 2022, Accepted 7 May 2022, Available online 16 May 2022, Version of Record 28 May 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117545