Score Bounded Monte-Carlo Tree Search

@inproceedings{Cazenave2010ScoreBM,
  title={Score Bounded Monte-Carlo Tree Search},
  author={Tristan Cazenave and Abdallah Saffidine},
  booktitle={Computers and Games},
  year={2010}
}
Monte-Carlo Tree Search (MCTS) is a successful algorithm used in many state of the art game engines. We propose to improve a MCTS solver when a game has more than two outcomes. It is for example the case in games that can end in draw positions. In this case it improves significantly a MCTS solver to take into account bounds on the possible scores of a node in order to select the nodes to explore. We apply our algorithm to solving Seki in the game of Go and to Connect Four. 

Figures, Tables, and Topics from this paper.

Citations

Publications citing this paper.
SHOWING 1-10 OF 24 CITATIONS

Opening Statistics and Match Play for Backgammon Games

VIEW 7 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Solving Games and All That

VIEW 5 EXCERPTS
CITES RESULTS, BACKGROUND & METHODS

Nash reweighting of Monte Carlo simulations: Tsumego

  • 2015 IEEE Congress on Evolutionary Computation (CEC)
  • 2015
VIEW 3 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

Simulation-Based General Game Playing

VIEW 2 EXCERPTS
CITES METHODS
HIGHLY INFLUENCED

Generalized Proof Number Search

VIEW 3 EXCERPTS
CITES RESULTS & METHODS

Strategic negotiations for extensive-form games

  • Autonomous Agents and Multi-Agent Systems
  • 2019
VIEW 1 EXCERPT
CITES METHODS

Pruning Game Tree by Rollouts

  • AAAI
  • 2015
VIEW 1 EXCERPT
CITES METHODS

References

Publications referenced by this paper.
SHOWING 1-8 OF 8 REFERENCES