Single-Player Monte-Carlo Tree Search

Maarten P. D. Schadd; Mark H. M. Winands; H. Jaap van den Herik; Guillaume M. J. -B. Chaslot; Jos W. H. M. Uiterwijk

doi:10.1007/978-3-540-87608-3_1

Single-Player Monte-Carlo Tree Search

Maarten P. D. Schadd^*
, Mark H. M. Winands
, H. Jaap van den Herik
, Guillaume M. J. -B. Chaslot
, Jos W. H. M. Uiterwijk

^*Corresponding author for this work

Networks and Strategic Optimization

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Academic

Abstract

Classical methods such as a* and ida* are a popular and successful choice for one-player games. However, they fail without an accurate admissible evaluation function. In this paper we investigate whether monte-carlo tree search (mcts) is an interesting alternative for one-player games where a* and ida* methods do not perform well. Therefore, we propose a new mcts variant, called single-player monte-carlo tree search (sp-mcts). The selection and backpropagation strategy in sp-mcts are different from standard mcts. Moreover, sp-mcts makes use of a straightforward meta-search extension. We tested the method on the puzzle samegame. It turned out that our sp-mcts program gained the highest score so far on the standardized test set.

Original language	English
Title of host publication	Computers and Games
Subtitle of host publication	6th International Conference, CG 2008, Beijing, China, September 29 - October 1, 2008. Proceedings
Place of Publication	Berlin, Heidelberg
Publisher	Springer
Pages	1-12
Number of pages	12
ISBN (Print)	978-3-540-87608-3
DOIs	https://doi.org/10.1007/978-3-540-87608-3_1
Publication status	Published - 2008

Publication series

Series	Lecture Notes in Computer Science
Volume	5131

Access to Document

10.1007/978-3-540-87608-3_1

http://dx.doi.org/10.1007/978-3-540-87608-3_1

Cite this

@inbook{81c51ea74817447c95c3a786bc014e4b,

title = "Single-Player Monte-Carlo Tree Search",

abstract = "Classical methods such as a* and ida* are a popular and successful choice for one-player games. However, they fail without an accurate admissible evaluation function. In this paper we investigate whether monte-carlo tree search (mcts) is an interesting alternative for one-player games where a* and ida* methods do not perform well. Therefore, we propose a new mcts variant, called single-player monte-carlo tree search (sp-mcts). The selection and backpropagation strategy in sp-mcts are different from standard mcts. Moreover, sp-mcts makes use of a straightforward meta-search extension. We tested the method on the puzzle samegame. It turned out that our sp-mcts program gained the highest score so far on the standardized test set.",

author = "Schadd, \{Maarten P. D.\} and Winands, \{Mark H. M.\} and \{van den Herik\}, \{H. Jaap\} and Chaslot, \{Guillaume M. J. -B.\} and Uiterwijk, \{Jos W. H. M.\}",

year = "2008",

doi = "10.1007/978-3-540-87608-3\_1",

language = "English",

isbn = "978-3-540-87608-3",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "1--12",

booktitle = "Computers and Games",

address = "United States",

}

Schadd, MPD, Winands, MHM, van den Herik, HJ, Chaslot, GMJ-B & Uiterwijk, JWHM 2008, Single-Player Monte-Carlo Tree Search. in Computers and Games: 6th International Conference, CG 2008, Beijing, China, September 29 - October 1, 2008. Proceedings. Springer, Berlin, Heidelberg, Lecture Notes in Computer Science, vol. 5131, pp. 1-12. https://doi.org/10.1007/978-3-540-87608-3_1

Single-Player Monte-Carlo Tree Search. / Schadd, Maarten P. D.; Winands, Mark H. M.; van den Herik, H. Jaap et al.
Computers and Games: 6th International Conference, CG 2008, Beijing, China, September 29 - October 1, 2008. Proceedings. Berlin, Heidelberg: Springer, 2008. p. 1-12 (Lecture Notes in Computer Science, Vol. 5131).

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Academic

TY - CHAP

T1 - Single-Player Monte-Carlo Tree Search

AU - Schadd, Maarten P. D.

AU - Winands, Mark H. M.

AU - van den Herik, H. Jaap

AU - Chaslot, Guillaume M. J. -B.

AU - Uiterwijk, Jos W. H. M.

PY - 2008

Y1 - 2008

N2 - Classical methods such as a* and ida* are a popular and successful choice for one-player games. However, they fail without an accurate admissible evaluation function. In this paper we investigate whether monte-carlo tree search (mcts) is an interesting alternative for one-player games where a* and ida* methods do not perform well. Therefore, we propose a new mcts variant, called single-player monte-carlo tree search (sp-mcts). The selection and backpropagation strategy in sp-mcts are different from standard mcts. Moreover, sp-mcts makes use of a straightforward meta-search extension. We tested the method on the puzzle samegame. It turned out that our sp-mcts program gained the highest score so far on the standardized test set.

AB - Classical methods such as a* and ida* are a popular and successful choice for one-player games. However, they fail without an accurate admissible evaluation function. In this paper we investigate whether monte-carlo tree search (mcts) is an interesting alternative for one-player games where a* and ida* methods do not perform well. Therefore, we propose a new mcts variant, called single-player monte-carlo tree search (sp-mcts). The selection and backpropagation strategy in sp-mcts are different from standard mcts. Moreover, sp-mcts makes use of a straightforward meta-search extension. We tested the method on the puzzle samegame. It turned out that our sp-mcts program gained the highest score so far on the standardized test set.

U2 - 10.1007/978-3-540-87608-3_1

DO - 10.1007/978-3-540-87608-3_1

M3 - Chapter

SN - 978-3-540-87608-3

T3 - Lecture Notes in Computer Science

SP - 1

EP - 12

BT - Computers and Games

PB - Springer

CY - Berlin, Heidelberg

ER -

Single-Player Monte-Carlo Tree Search

Abstract

Publication series

Access to Document

Fingerprint

Cite this