An analysis of diversity of constants of Genetic Programming

Conor Ryan, Maarten Keijzer

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

Abstract

This paper conducts an investigation into the manner in which constants evolve during the course of GP run. It starts by describing an intuitive Gaussian type mutation for constants and showing that its ability to produce small changes in individuals leads to a high performance. It then demonstrates the surprising result that, in a selection of real world problems, simple random mutation performs better. The paper then finishes with an analysis of the diversity of constants in the population, and the manner in which this changes over time.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
EditorsConor Ryan, Terence Soule, Maarten Keijzer, Edward Tsang, Riccardo Poli, Ernesto Costa
PublisherSpringer Verlag
Pages404-413
Number of pages10
ISBN (Print)354000971X, 9783540009719
DOIs
Publication statusPublished - 2003

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2610
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Fingerprint

Dive into the research topics of 'An analysis of diversity of constants of Genetic Programming'. Together they form a unique fingerprint.

Cite this