DocumentCode
2031306
Title
A new hillclimber for classifier systems
Author
Tsui, Kwok Ching ; Plumbley, Mark
Author_Institution
Dept. of Comput. Sci., King´´s Coll., London, UK
fYear
1997
fDate
2-4 Sep 1997
Firstpage
97
Lastpage
102
Abstract
Multistate artificial environments such as mazes represent a class of tasks that can be solved by many different multistep methods. When different rewards are available in different places of the maze, a problem solver is required to evaluate different positions effectively and remembers the best one. A new hillclimbing strategy for the Michigan style classifier system is suggested which is able to find the shortest path and discarding suboptimal solutions. Knowledge reuse is also shown to be possible
Keywords
genetic algorithms; Michigan style classifier system; genetic algorithm; hillclimber; knowledge reuse; mazes; multistate artificial environments; problem solver; shortest path; suboptimal solution discarding;
fLanguage
English
Publisher
iet
Conference_Titel
Genetic Algorithms in Engineering Systems: Innovations and Applications, 1997. GALESIA 97. Second International Conference On (Conf. Publ. No. 446)
Conference_Location
Glasgow
ISSN
0537-9989
Print_ISBN
0-85296-693-8
Type
conf
DOI
10.1049/cp:19971162
Filename
680990
Link To Document