s'authentifier
version française rss feed
inria-00117266, version 3
Voir la fiche détaillée  BibTeX  EndNote  TEI  RefWorks
Modification of UCT with Patterns in Monte-Carlo Go
Sylvain Gelly (, http://www.lri.fr/~gelly/) 1, Yizao Wang 12, Rémi Munos 23, Olivier Teytaud 1
(2006)
Icone de RR-6062.pdf
Algorithm UCB1 for multi-armed bandit problem has already been extended to Algorithm UCT (Upper bound Confidence for Tree) which works for minimax tree search. We have developed a Monte-Carlo Go program, MoGo, which is the first computer Go program using UCT. We explain our modification of UCT for Go application and also the intelligent random simulation with patterns which has improved significantly the performance of MoGo. UCT combined with pruning techniques for large Go board is discussed, as well as parallelization of UCT. MoGo is now a top level Go program on $9\times9$ and $13\times13$ Go boards.
1 :  TAO (INRIA Futurs)
INRIA – CNRS : UMR8623 – Université Paris Sud - Paris XI
2 :  Centre de Mathématiques Appliquées (CMAP)
CNRS : UMR7641 – Université de Versailles-Saint Quentin en Yvelines – Polytechnique - X
3 :  SEQUEL (INRIA Futurs)
INRIA – CNRS : UMR8022 – CNRS : UMR8146 – Université des Sciences et Technologies de Lille - Lille I – Université Charles de Gaulle - Lille III – Ecole Centrale de Lille
Informatique/Intelligence artificielle
Informatique/Apprentissage
RR-6062