PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Monte-Carlo Simulation Balancing in Practice
Shih-Chieh Huang, Rémi Coulom and Shun-Shii Lin
In: Computers and Games, Kanazawa, Japan(2010).


Simulation balancing is a new technique to tune parameters of a playout policy for a Monte-Carlo game-playing program. So far, this algorithm had only been tested in a very artificial setting: it was limited to $5\times5$ and $6\times6$ Go, and required a stronger external program that served as a supervisor. In this paper, the effectiveness of simulation balancing is demonstrated in a more realistic setting. A state-of-the-art program, \Erica, learned an improved playout policy on the $9\times9$ board, without requiring any external expert to provide position evaluations. Evaluations were collected by letting the program analyze positions by itself. The previous version of \Erica\ learned pattern weights with the minorization-maximization algorithm. Thanks to simulation balancing, its playing strength was improved from a winning rate of 69\% to 78\% against \Fuego~0.4.

EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:7423
Deposited By:Rémi Coulom
Deposited On:17 March 2011