• 検索結果がありません。

要 旨 オセロにおける モンテカルロ法プレイヤの性能調査

N/A
N/A
Protected

Academic year: 2021

シェア "要 旨 オセロにおける モンテカルロ法プレイヤの性能調査"

Copied!
2
0
0

読み込み中.... (全文を見る)

全文

(1)

要 旨

オセロにおける

モンテカルロ法プレイヤの性能調査

稲田 康平

モンテカルロ法は乱数を用いたシミュレーション (プレイアウト) を複数回行って近似解 を求める計算手法である.本研究ではモンテカルロ法を利用したオセロプレイヤを作成し,

乱数プレイヤと試合を行うことでモンテカルロ法プレイヤの性能を調査する.

本論文では,まずオセロとモンテカルロ法について説明する.そして,モンテカルロ法プ レイヤの評価値の取り方やプレイアウト数を変化させることで,モンテカルロ法プレイヤの 性能がどのように変化するかを調査する.またモンテカルロ法プレイヤが最善手を選べてい るかどうかについても調査する.

調査の結果,評価の取り方に関わらずプレイアウト数を増やすほどモンテカルロ法プレイ ヤの性能が良くなることが分かった.またモンテカルロ法プレイヤは最善手ではなく,悪手 が少なくなるように手を選んでいることが分かった.

キーワード オセロ (リバーシ) ,モンテカルロ法

i

(2)

Abstract

The performance investigation of the Monte-Carlo player in othello

Kohei INATA

The Monte Carlo method is an algorithm that computes an approximate solution by a number of random simulations (playouts) . In this study, an othello player based on the Monte Carlo method is created, and the performance of the Monte-Carlo player is investigated with respect to random-play player.

This thesis first explains the othello game and the Monte Carlo method. Then, we investigate the performance of the Monte-Carlo player by changing the method of computing evaluation values and the number of playouts. Moreover, we also investigate whether the Monte-Carlo player chooses the best move.

The results show the followings. For any method of computing evaluation values, the Monte-Carlo player shows the better performance when the more number of playouts are performed. It also turns out that the Monte-Carlo player chooses not the best move but a move that has less possibility to become a bad move.

key words Othello (Reversi) , Monte Carlo Method

ii

参照

関連したドキュメント

Figure 6: To the left, the upper P-positions of Maharaja Nim in columns 8 to 12 have been computed, beginning with position (8, 13), and a perfect sector has been detected.. The

In general, SDEs under regime-switching have no explicit solutions so the Monte Carlo simulations have become one of the powerful techniques in valuation of financial quantities,

Then, in the middle we illustrate Wythoff Nim’s pair of P-beams with slopes φ and 1/φ respectively and, at last, we present the initial P-positions of (1, 2)GDWN, where our

The goal of this work is to study the performance of the estimates produced by the EM algorithm, taking into account the method of moments and a random initialization method to

In Section 3 we study the current time correlations for stationary lattice gases and in Section 4 we report on Monte-Carlo simulations of the TASEP in support of our

Furuta, Log majorization via an order preserving operator inequality, Linear Algebra Appl.. Furuta, Operator functions on chaotic order involving order preserving operator

In other words, the aggressive coarsening based on generalized aggregations is balanced by massive smoothing, and the resulting method is optimal in the following sense: for

By an inverse problem we mean the problem of parameter identification, that means we try to determine some of the unknown values of the model parameters according to measurements in