Abstract. We investigate the use of Monte-Carlo Tree Search (MCTS) within the field of computer Poker, more specifically No-Limit Texas Hold’em. The hidden information in Poker results in so called miximax game trees where opponent decision nodes have to be modeled as chance nodes. The probability distribution in these nodes is modeled by an opponent model that predicts the actions of the opponents. We propose a modification of the standard MCTS selection and backpropagation strategies that explicitly model and exploit the uncertainty of sampled expected values. The new strategies are evaluated as a part of a complete Poker bot that is, to the best of our knowledge, the first exploiting no-limit Texas Hold’em bot that can play at a reasonable level in games of more than two players. 1

@inproceedings{VdBACML09,
  author = "Van den Broeck, Guy and Driessens, Kurt and Ramon, Jan",
  title = "Monte-{C}arlo tree search in poker using expected reward distributions",
  booktitle = "Proceedings of the 1st Asian Conference on Machine Learning (ACML), Lecture Notes in Computer Science",
  publisher = "Springer",
  pages = "367--381",
  year = "2009",
  doi = "10.1007/978-3-642-05224-8",
  month = 8,
  url="http://starai.cs.ucla.edu/papers/VdBACML09.pdf",
  keywords   = {conference,selective}
}