SAint7579 / Monte-Carlo-agent-for-TicTacToe

An RL agent that plays tic tac toe using epsilon greedy selection and monte carlo updates

Monte Carlo based RL agent for Tic Tac Toe

An RL agent that plays tic tac toe using epsilon greedy selection and monte carlo updates.
The training is done using self-play and over 20000 episodes of the game.

Libraries used:

• Numpy

Screenshot: