Sequential halving for partially observable games

Tom Pepels; Tristan Cazenave; Mark H M Winands

doi:10.1007/978-3-319-39402-2_2

Sequential halving for partially observable games

Tom Pepels^*, Tristan Cazenave, Mark H M Winands

^*Corresponding author for this work

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

Abstract

This paper investigates Sequential Halving as a selection policy in the following four partially observable games: Go Fish, Lost Cities, Phantom Domineering, and Phantom Go. Additionally, H-MCTS is studied, which uses Sequential Halving at the root of the search tree, and UCB elsewhere. Experimental results reveal that H-MCTS performs the best in Go Fish, whereas its performance is on par in Lost Cities and Phantom Domineering. Sequential Halving as a at Monte-Carlo Search appears to be the stronger technique in Phantom Go.

Original language	English
Title of host publication	Computer Games
Publisher	Springer
Pages	16-29
Number of pages	14
ISBN (Print)	9783319394015
DOIs	https://doi.org/10.1007/978-3-319-39402-2_2
Publication status	Published - 2016

Publication series

Series	Communications in Computer and Information Science
Volume	614

Access to Document

10.1007/978-3-319-39402-2_2

http://www.mendeley.com/research/sequential-halving-partially-observable-games

Cite this

@inbook{313fe98055e74d82a0f77f2c36e94378,

title = "Sequential halving for partially observable games",

abstract = "This paper investigates Sequential Halving as a selection policy in the following four partially observable games: Go Fish, Lost Cities, Phantom Domineering, and Phantom Go. Additionally, H-MCTS is studied, which uses Sequential Halving at the root of the search tree, and UCB elsewhere. Experimental results reveal that H-MCTS performs the best in Go Fish, whereas its performance is on par in Lost Cities and Phantom Domineering. Sequential Halving as a at Monte-Carlo Search appears to be the stronger technique in Phantom Go.",

author = "Tom Pepels and Tristan Cazenave and Winands, {Mark H M}",

year = "2016",

doi = "10.1007/978-3-319-39402-2_2",

language = "English",

isbn = "9783319394015",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "16--29",

booktitle = "Computer Games",

address = "United States",

}

TY - CHAP

T1 - Sequential halving for partially observable games

AU - Pepels, Tom

AU - Cazenave, Tristan

AU - Winands, Mark H M

PY - 2016

Y1 - 2016

N2 - This paper investigates Sequential Halving as a selection policy in the following four partially observable games: Go Fish, Lost Cities, Phantom Domineering, and Phantom Go. Additionally, H-MCTS is studied, which uses Sequential Halving at the root of the search tree, and UCB elsewhere. Experimental results reveal that H-MCTS performs the best in Go Fish, whereas its performance is on par in Lost Cities and Phantom Domineering. Sequential Halving as a at Monte-Carlo Search appears to be the stronger technique in Phantom Go.

AB - This paper investigates Sequential Halving as a selection policy in the following four partially observable games: Go Fish, Lost Cities, Phantom Domineering, and Phantom Go. Additionally, H-MCTS is studied, which uses Sequential Halving at the root of the search tree, and UCB elsewhere. Experimental results reveal that H-MCTS performs the best in Go Fish, whereas its performance is on par in Lost Cities and Phantom Domineering. Sequential Halving as a at Monte-Carlo Search appears to be the stronger technique in Phantom Go.

U2 - 10.1007/978-3-319-39402-2_2

DO - 10.1007/978-3-319-39402-2_2

M3 - Chapter

SN - 9783319394015

T3 - Communications in Computer and Information Science

SP - 16

EP - 29

BT - Computer Games

PB - Springer

ER -