On the Huge Benefit of Decisive Moves in Monte-Carlo Tree Search Algorithms

Fabien Teytaud 1, 2 Olivier Teytaud 1, 2
1 TAO - Machine Learning and Optimisation
LRI - Laboratoire de Recherche en Informatique, UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France, CNRS - Centre National de la Recherche Scientifique : UMR8623
Abstract : Monte-Carlo Tree Search (MCTS) algorithms, including upper confidence Bounds (UCT), have very good results in the most difficult board games, in particular the game of Go. More recently these methods have been successfully introduce in the games of Hex and Havannah. In this paper we will define decisive and anti-decisive moves and show their low computational overhead and high efficiency in MCTS.
Document type :
Conference papers
Complete list of metadatas

Cited literature [32 references]  Display  Hide  Download

Contributor : Fabien Teytaud <>
Submitted on : Friday, June 25, 2010 - 8:50:33 AM
Last modification on : Thursday, April 5, 2018 - 12:30:12 PM
Long-term archiving on : Monday, September 27, 2010 - 11:55:43 AM


Files produced by the author(s)


  • HAL Id : inria-00495078, version 1



Fabien Teytaud, Olivier Teytaud. On the Huge Benefit of Decisive Moves in Monte-Carlo Tree Search Algorithms. IEEE Conference on Computational Intelligence and Games, Aug 2010, Copenhagen, Denmark. ⟨inria-00495078⟩



Record views


Files downloads