An Evaluation Function for Othello Based on Statistics
نویسنده
چکیده
This paper describes the evaluation function of one of today's strongest Othello programs | LOGISTELLO. The function is based on statistical analysis of a large set of example game positions and proots from ideas regarding pattern value estimation, determination of feature weights, and the fast approximation of time consuming evaluation features.
منابع مشابه
Strategy Acquisition for the Game "Othello" Based on Reinforcement Learning
This article discusses automatic strategy acquisition for the game \Othello" based on reinforcement learning. In our approach, two computer players initially know only the game rules, but they become relatively stronger after playing several thousands of games against each other. In each game, the players re ne the evaluation function for the game state, which is achieved in a reinforcement lea...
متن کاملExperiments with Multi{ProbCut and a New High{Quality Evaluation Function for Othello
This paper presents ideas concerning game{tree evaluation that recently improved the author's strong Othello program LOGISTELLO considerably. Two main parts of this program have been reened. In the rst part of this paper, a new evaluation function for Othello is described. While it is table{based like its predecessors , a novel procedure for estimating table entries now allows modeling pattern ...
متن کاملBILL : a table-based, knowledge-intensive othello program
A constant dilemma facing game-playing programs is whether to emphasize searching or knowledge. This paper describes a world-championship level Othello program, BILL, that succeeds in both dimensions. The success of BILL is largely due to its understanding of many important Othello features by using a pre-compiled knowledge base of board patterns. Because of this pre-compiled nature of its know...
متن کاملTeaching the HONEST Neural Network to Judge the Strength of Othello Positions
Bayesian learning, a statistical method, has been successfully applied to the feature combination stage of the evaluation function of the Othello program BILL, which won the 1989 North American Computer World Othello Championship. We have investigated the use of a novel neural network called HONEST in place of Bayesian learning in BILL. We implemented HONEST in six of BILL's twenty-six evaluati...
متن کاملImproving heuristic mini-max search by supervised learning
This article surveys three techniques for enhancing heuristic game-tree search pioneered in the author’s Othello program LOGISTELLO, which dominated the computer Othello scene for several years and won against the human World-champion 6–0 in 1997. First, a generalized linear evaluation model (GLEM) is described that combines conjunctions of Boolean features linearly. This approach allows an aut...
متن کامل