Date
2016
Dewey
Méthodes informatiques spéciales
Sujet
Monte Carlo Tree Search
Book title
Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (AAAI-16)
Author
Dale Schuurmans, Michael Wellman
Publisher
AAAI Press
Publisher city
Palo Alto (USA)
Year
2016
ISBN
978-1-57735-760-5
Author
Cazenave, Tristan
Saffidine, Abdallah
Schofield, Michael John
Thielscher, Michael
Type
Communication / Conférence
Item number of pages
687-693
Abstract (EN)
The use of the Monte Carlo playouts as an evaluation function has proved to be a viable, general technique for searching intractable game spaces. This facilitate the use of statistical techniques like Monte Carlo Tree Search (MCTS), but is also known to require significant processing overhead. Weseek to improve the quality of information extracted from the Monte Carlo playout in three ways. Firstly, by nesting the evaluation function inside another evaluation function; secondly, by measuring and utilising the depth of the playout; and thirdly, by incorporating pruning strategies that eliminateunnecessary searches and avoid traps. Our experimental data, obtained on a variety of two-player games from past General Game Playing (GGP) competitions and others, demonstrate the usefulness of these techniques in a Nested Player when pitted against a standard, optimised UCT player.