Combining UCT and Nested Monte Carlo Search for Single-Player General Game Playing

Monte Carlo tree search (MCTS) has been recently very successful for game playing, particularly for games where the evaluation of a state is difficult to compute, such as Go or General Games. We compare nested Monte Carlo (NMC) search, upper confidence bounds for trees (UCT-T), UCT with transpositio...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on computational intelligence and AI in games. Vol. 2; no. 4; pp. 271 - 277
Main Authors: Méhat, Jean, Cazenave, Tristan
Format: Journal Article
Language:English
Published: IEEE 01.12.2010
Subjects:
Algorithm design and analysis
Artificial intelligence
Classification algorithms
Competition
Computer simulation
Decision trees
Games
General Game Playing (GPP)
Monte Carlo methods
nested Monte-Carlo search
Searching
single-player games
Tables
Trees
upper confidence bounds for trees (UCT)
ISSN:1943-068X, 1943-0698
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Monte Carlo tree search (MCTS) has been recently very successful for game playing, particularly for games where the evaluation of a state is difficult to compute, such as Go or General Games. We compare nested Monte Carlo (NMC) search, upper confidence bounds for trees (UCT-T), UCT with transposition tables (UCT+T), and a simple combination of NMC and UCT+T (MAX) on single-player games of the past General Game Playing (GGP) competitions. We show that transposition tables improve UCT and that MAX is the best of these four algorithms. Using UCT+T, the program Ary won the 2009 GGP competition. MAX and NMC are slight improvements over this 2009 version.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:1943-068X
1943-0698
DOI:10.1109/TCIAIG.2010.2088123