An Q-Learning agent that learns to play 3x3 tic-tac-toe
board.py -> class Board; player.py -> class Player; learn.py -> the learning process, run this code to generate strategies: use Q_1 if AI player starts first, use Q_2 if AI player has to start second; game.py -> main code for game to play against human; NN_Q.py -> keras based neural net modeled to replace Q table (work in progress)