Multi-Subexpression Programming

Gene Expression Programming is a new and adaptive brand evolution algorithm which is developed on the basis of genetic algorithm. In recent years, Multi-Expression Programming which is proposed in the genetic programming is a linear structure coding scheme,its main feature is a chromosome contains m...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Applied Mechanics and Materials Ročník 411-414; s. 2067 - 2073
Hlavní autori: Chen, Long Bin, He, Pei
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Zurich Trans Tech Publications Ltd 01.09.2013
Predmet:
Average Fitness Value
MEP
GEP
Success Rate
Subexpression
Evolutionary Efficiency
ISBN:3037858648, 9783037858646
ISSN:1660-9336, 1662-7482, 1662-7482
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Gene Expression Programming is a new and adaptive brand evolution algorithm which is developed on the basis of genetic algorithm. In recent years, Multi-Expression Programming which is proposed in the genetic programming is a linear structure coding scheme,its main feature is a chromosome contains multiple expressions. The idea of MEP is introduced into the GEP in this paper, so a single GEP gene contains multiple solutions to solve the problem.The new algorithm analyzes each gene in the GEP to extract relational subexpressions, then fitness evaluate certain subexpressions to choose the best fitness as individuals fitness, and carry on related genetic manipulation. Finally, the improved algorithm experiment with GEP and MEP, compare their mining the same functions ability,record average fitness value and success rate. The experiment results show that the improved algorithm has better evolutionary efficiency.
Bibliografia:Selected, peer reviewed papers from the 2013 2nd International Conference on Information Technology and Management Innovation (ICITMI 2013), July 23-24, 2013, Zhuhai, China
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISBN:3037858648
9783037858646
ISSN:1660-9336
1662-7482
1662-7482
DOI:10.4028/www.scientific.net/AMM.411-414.2067