Genetic Programming

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

1 Citation (Scopus)


GAs, studied in Chap. 3, are capable of solving many problems and simple enough to allow for solid theoretical studies. Nevertheless, the representation of individuals that characterizes GAs (i.e., the fact that individuals in GAs must be strings of a previously fixed length) can be a limitation for a wide set of applications. In these cases, the most natural representation for a solution is a hierarchical computer program, rather than a string of characters of a fixed length. For example, strings of a static length do not readily support the hierarchical organization of tasks into subtasks typical of computer programs, they do not provide any convenient way of incorporating iteration and recursion, and so on. But above all, GA representation schemes do not have any dynamic variability: the initial selection of string length limits in advance the number of internal states of the system and limits what the system can learn.

Original languageEnglish
Title of host publicationLectures on Intelligent Systems
Place of PublicationCham, Switzerland
PublisherSpringer, Cham
Number of pages53
ISBN (Electronic)978-3-031-17922-8
ISBN (Print)978-3-031-17921-1, 978-3-031-17924-2
Publication statusPublished - 13 Jan 2023

Publication series

NameNatural Computing Series
ISSN (Print)1619-7127


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

Cite this