Skip to main navigation Skip to search Skip to main content

Exploring Non-bloating Geometric Semantic Genetic Programming

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

1 Downloads (Pure)

Abstract

Recently, a new variant of Geometric Semantic Genetic Programming(GSGP) was introduced that, while maintaining the property of inducing a unimodal error surface for all supervised learning problems, is able to generate models that are compact enough to be interpretable by humans. This variant is called the Semantic Learning algorithm based on Inflate and deflate Mutation (SLIM_GSGP) and, as the name suggests, it is based on two types of mutation: one (inflate) that generates offspring larger than their parents, similar to traditional geometric semantic mutation, and the other (deflate) that generates offspring smaller than their parents. This chapter reviews and extends the initial work on SLIM_GSGP by introducing two novel variants, thoroughly studying the geometric characteristics of the SLIM_GSGP operators and discussing the experimental results and their interpretation in greater depth. The main conclusion is that SLIM_GSGP is a very promising method, warranting significant investment in future research.
Original languageEnglish
Title of host publicationGenetic Programming Theory and Practice XXI
EditorsStephan M. Winkler, Wolfgang Banzhaf, Ting Hu, Alexander Lalejini
Place of PublicationSingapore
PublisherSpringer Singapore
Pages237-258
Number of pages22
ISBN (Electronic)978-981-96-0077-9
ISBN (Print)978-981-96-0076-2
DOIs
Publication statusPublished - 28 Feb 2025

Publication series

NameGenetic and Evolutionary Computation
PublisherSpringer Singapore
ISSN (Print)1932-0167
ISSN (Electronic)1932-0175

Fingerprint

Dive into the research topics of 'Exploring Non-bloating Geometric Semantic Genetic Programming'. Together they form a unique fingerprint.

Cite this