Multipurpose strategic planning in the game of Go

被引:3
作者
Hu, S [1 ]
Lehner, PE [1 ]
机构
[1] MITRE CORP,MCLEAN,VA 22101
关键词
computer Go; adversarial planning; automated planning; artificial intelligence;
D O I
10.1109/34.615454
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
A heuristic method for adversarial planning is developed to address the problem of multipurpose planning in the game of Go. Static analysis and dynamic look ahead on both strategic and tactical levels are used to generate possible goals and to identify interactions among the achievability of various goals. Strategic, multipurpose goals are composed of sets of interacting goals.
引用
收藏
页码:1048 / 1051
页数:4
相关论文
共 18 条
[1]  
[Anonymous], [No title captured]
[2]  
[Anonymous], P 6 INT JOINT C ART
[3]  
BENSON DB, 1976, INFORM SCIENCES, V10, P17, DOI 10.1016/0020-0255(76)90059-1
[4]  
CHEN K, 1990, COMPUTER CHESS COGNI, P271
[5]  
CHEN K, 1989, HEURISTIC PROGRAMMIN, P195
[6]  
FRIEDENBACH KJ, 1980, THESIS U CALIFORNIA
[7]  
HU S, 1995, THESIS G MASON U
[8]  
KAGEYAMA T, 1978, LESSONS FUNDAMENTALS
[9]  
KIERULF A, 1990, COMM ACM FEB, P152
[10]  
LEHNER PE, 1990, ADV TECHNOLOGY COMMA