Understanding grammatical evolution: initialisation

Grammatical evolution is one of the most used variants of genetic programming, and ever since its introduction, several improvements have been suggested. One of these concerns the routine used to create the initial population. In this study, several proposed initialisation routines are compared; bas...

Full description

Saved in:
Bibliographic Details
Published inGenetic programming and evolvable machines Vol. 18; no. 4; pp. 467 - 507
Main Author Nicolau, Miguel
Format Journal Article
LanguageEnglish
Published New York Springer US 01.12.2017
Springer Nature B.V
Subjects
Artificial Intelligence
Biological evolution
Biomedical Engineering and Bioengineering
Compilers
Computer Science
Electrical Engineering
Genetic algorithms
Interpreters
Programming Languages
Programming Techniques
Routines
Software Engineering/Programming and Operating Systems
Representation bias
Design
Symbolic regression
Grammatical evolution
Classification
Initialisation
Tree creation
Online AccessGet full text
ISSN1389-2576
1573-7632
DOI10.1007/s10710-017-9309-9

Cover

More Information
Summary:Grammatical evolution is one of the most used variants of genetic programming, and ever since its introduction, several improvements have been suggested. One of these concerns the routine used to create the initial population. In this study, several proposed initialisation routines are compared; based on a detailed analysis of the generated initial populations, and subsequent results obtained on a large set of experiments, a variant of the PTC2 algorithm is shown to consistently outperform all other routines, while a variant of random initialisation provides a good compromise between efficiency and ease of implementation.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1389-2576
1573-7632
DOI:10.1007/s10710-017-9309-9