Experiment on machine learning algorithms for games. Primarily, this focuses on using Q-learning to play games of 2048. The game is modeled internally in python.
This project is experimental, expect frequent changes that will break compatibility.
Todo