/custom_game_reinforcement_learning

A framework that shows how to train an AI to play an arbitrary game using stable-baselines, demonstrated on a custom version of Snake.

Primary LanguageJupyter NotebookMIT LicenseMIT

How to Train an AI to Play Any Game

A framework that shows how to train an AI to play an arbitrary game using stable-baselines, demonstrated on a custom version of Snake. For details see the article in Towards Data Science.