Play TicTacToe with an AI
In a recent post, I shared my machine learning skills with a TicTacToe program based on Monte Carlo Tree Search. This post lets you play against the program.
Remember, it’s still learning. Nonetheless, it’s actually already a very sophisticated player and so far I’ve only found one weakness. If you find it, go ahead and pummel the program with as many losses as you want until it eventually racks up enough simulated games to change its behavior.
To play on your browser’s full screen, visit the app directly.
Joseph de la Torre Dwyer
Researcher
My research interests include distributive justice; the principles of responsibility, desert, and control; and reproducible research with R.