How to use crowding selection in grammar-based classifier system

Author

Unold, Olgierd ; Cielecki, Lukasz

Author_Institution

Inst. of Eng. Cybern., Wroclaw Univ. of Technol., Poland

fYear

2005

fDate

8-10 Sept. 2005

Firstpage

124

Lastpage

129

Abstract

The grammar-based classifier system (GCS) is a new version of learning classifier systems (LCS) in which classifiers are represented by context-free grammar in Chomsky normal form. GCS evolves one grammar during induction (the Michigan approach) which gives it the ability to find the proper set of rules very quickly. However it is quite sensitive to any variations of learning parameters. This paper investigates the role of crowding selection in GCS. To evaluate the performance of GCS depending on crowding factor and crowding subpopulation we used context-free language in the form of so-called toy language. The set of experiments was performed to obtain the answer for question in the title.

Keywords

context-free grammars; context-free languages; inference mechanisms; learning (artificial intelligence); pattern classification; Chomsky normal form; Michigan approach; context-free grammar; context-free language; crowding factor; crowding selection; crowding subpopulation; grammar-based classifier system; learning classifier system; toy language; Cybernetics; Detectors; Genetic algorithms; Intelligent systems; Natural languages; Polynomials; Production systems; Protection; Trademarks;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Systems Design and Applications, 2005. ISDA '05. Proceedings. 5th International Conference on

Print_ISBN

0-7695-2286-6

Type

conf

DOI

10.1109/ISDA.2005.50

Filename

1578772