Applying Reinforcement Learning to Blackjack
The game of Blackjack is a popular card game played in various casinos. It is a comparing card game between a dealer and multiple players but players are independent with each other. The object of the game is to win money by obtaining a sum of card values that is equal to or less than 21 so that their total values are higher than the dealer's. In this paper, I proposed three different reinforcement algorithms as blackjack strategies and the AI learns how to play to get more rewards. I implement the blackjack game and reinforcement algorithms using Python. Furthermore, I compare the performance among these three strategies (Monte Carlo method, Q-learning method, Sarsa method).