Designed and implemented an online-poker playing robot. The base algorithm compared the pot odds to the probabilistic likelihood of winning the hand with the information available. Using a Monte Carlo simulation method, I guess the future community cards to determine my possibility of winning.
Machine learning is then used over time to feedback information on other players' behavior to make "my" playing more intelligent.
Used the Java "Robot" class to simulate mouse movements and mouse clicks in order to interface with Full Tilt Poker, and play real people (or other bots) online.
Link to this comment:
All Comments (0)