/TicTacToe

A Reinforcement Learning approach to the general TicTacToe games

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

Generalized TicTacToe

Who does not know the Tic Tac Toe? This notebook aims at mastering the N*N tic tac toe, with N >= 3 and use it as a practical example of the benefits of reinforcement learning.

Then, the goal is to master a childhood game. The rules are simple: each player scores a point for 5 aligned crosses/rounds and the highest score wins when no more spots are available (or meaningful). Any grid size can work, but typically 30 by 30 is used.