q-p Yet another implementation of tic-tac-toe game using Q-learning. Usage (let [{:keys [x o]} (train q-player q-player 200000)] (play x o)) Next steps Implement neural network to be able to handle larger grids & more complex states generally.