%0 Conference Proceedings
%T Comparative Study of Monte-Carlo Tree Search and Alpha-Beta Pruning in Amazons
%+ Akita University
%A Kato, Hikari
%A Fazekas, Szilárd, Zsolt
%A Takaya, Mayumi
%A Yamamura, Akihiro
%Z Part 5: Applied Modeling and Simulation
%< avec comité de lecture
%( Lecture Notes in Computer Science
%B 3rd International Conference on Information and Communication Technology-EurAsia (ICT-EURASIA) and 9th International Conference on Research and Practical Issues of Enterprise Information Systems (CONFENIS)
%C Daejon, South Korea
%Y Ismail Khalil
%Y Erich Neuhold
%Y A Min Tjoa
%Y Li Da Xu
%Y Ilsun You
%I Springer
%3 Information and Communication Technology
%V LNCS-9357
%P 139-148
%8 2015-10-04
%D 2015
%R 10.1007/978-3-319-24315-3_14
%K Amazons
%K Two player games
%K Monte-Carlo tree search
%K UCT
%K Game programs
%K Playouts
%K Alpha-Beta pruning
%K Evaluation function
%Z Computer Science [cs]
%Z Humanities and Social Sciences/Library and information sciencesConference papers
%X The game of Amazons is a combinatorial game sharing some properties of both chess and Go.We study programs which play Amazons with strategies based on Monte-Carlo Tree Search and a classical search algorithm, Alpha-Beta pruning.We execute several experiments to investigate the effect of increasing the number of searches in a Monte-Carlo Tree Search program. We show that increasing the number of searches is not an efficient method to strengthen the program for Amazons. On the other hand, augmenting the algorithms with a choice of several evaluation functions fulfills has great influence on playing strength.
%G English
%Z TC 5
%Z TC 8
%Z WG 8.9
%2 https://inria.hal.science/hal-01466213/document
%2 https://inria.hal.science/hal-01466213/file/978-3-319-24315-3_14_Chapter.pdf
%L hal-01466213
%U https://inria.hal.science/hal-01466213
%~ SHS
%~ IFIP-LNCS
%~ IFIP
%~ IFIP-TC
%~ IFIP-TC5
%~ IFIP-WG
%~ IFIP-TC8
%~ IFIP-ICT-EURASIA
%~ IFIP-WG8-9
%~ IFIP-LNCS-9357
%~ IFIP-CONFENIS