Multi-chromosomal Grammatical Evolution

Hara Akira

Yamaguchi Tomohisa

Ichimura Takumi

Takahama Tetsuyuki

Fourth International Workshop on Computational Intelligence & Applications Proceedings : IWCIA 2008

Grammatical Evolution (GE) is an evolutionary method for optimizing a program generated by a one-dimensional chromosome and grammatical rules. The grammars consist of terminals, which are items that can appear in the language, and nonterminals, which can be expanded into one or more terminals and nonterminals. The genes are translated into a program based on the grammar. If the genes are used up for generating complete program, the chromosome is wrapped and reused. GE has an advantage that illegal individuals are not generated by the genetic operations. When a certain gene changes, however, the successive genes might be used for the different production rule from the rule applied before even if they are not changed. Therefore, it is difficult to preserve the characteristics of parents. To solve this problem, we propose GE using multiple chromosomes. In this method, multiple chromosomes as many as the nonterminal symbols in the grammatical rules are prepared. A chromosome correspondent to the expanded non-terminal symbol is selected and used for mapping. Moreover, a new technique of the wrapping is also introduced so that the grammatical rules which increase the number of nonterminal symbols can not be applied when the wrapping happens. We performed some experiments, and showed the effectiveness of our proposed method.

技術・工学 [ 500 ]

英語

IEEE SMC Hiroshima Chapter

(c) Copyright by IEEE SMC Hiroshima Chapter.

[ISSN] 1883-3977