Using CNN for solving two-player zero-sum games
Résumé
We study a two-player zero-sum game (matrix game for short) with the objective to find the saddle point and its value. We develop a novel convolutional neural network (CNN for short) approach to achieve the goal. We propose a complete training pipeline, including a specific CNN model structure to handle varying game size, generating training dataset and model fitting. The experiment results show that our proposed method outperforms the traditional linear programming (LP for short) method in terms of computational efforts.
Origine : Fichiers produits par l'(les) auteur(s)