Export Ready — 

Parameter Control in Geometric Semantic Genetic Programming

Bibliographic Details
Main Author: Dinis, Marta Teixeira Rumina
Publication Date: 2024
Format: Master thesis
Language: eng
Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
Download full: http://hdl.handle.net/10362/174532
Summary: Dissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Data Science
id RCAP_f1280d16befa2bd3540b8dfa71bbdb31
oai_identifier_str oai:run.unl.pt:10362/174532
network_acronym_str RCAP
network_name_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository_id_str https://opendoar.ac.uk/repository/7160
spelling Parameter Control in Geometric Semantic Genetic ProgrammingGenetic ProgrammingGeometric Semantic Genetic ProgrammingGenetic OperatorsMutationParameter ControlDomínio/Área Científica::Ciências Naturais::Ciências da Computação e da InformaçãoDissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Data ScienceMachine Learning (ML) is a scientific field that aims to enable computers to learn without explicit programming. Evolutionary Algorithms (EAs), a subset of ML, mimic natural selection to evolve solutions to problems. Among EAs, Genetic Programming (GP) is one of the latest advancements, evolving computer programs to accurately transform input data into expected outputs. Moraglio et al. (2012) first introduced the Geometric Semantic Genetic Programming (GSGP) algorithm to improve upon the GP algorithm by applying genetic operators that have a known e!ect on the semantic space of the individuals. This results on a unimodal fitness landscape that allows the algorithm to have very good optimization ability. However, GSGP is not without its problems. Particularly, the usage of parameters that are shared by all individuals and remain constant throughout the evolution may inhibit or delay the obtainment of the optimal solution. In this work, the application of parameter control methods to the mutation step parameter within the Geometric Semantic Mutation (GSM) operator is proposed with the objective to solve this issue and improve GSGP’s performance. The core of this project is the creation of four di!erent mutation step variants to the standard GSGP. Other adjustments had to be made in order to implement these variants seamlessly. Particularly, the mutation step is no longer a feature of the whole population, but of each single individual, and its value is bounded by values dependent on the fitness of the individual in question. The performance of the proposed method was tested on nine regression benchmark datasets. The tests conducted show promising results for the application of parameter control to the mutation step, generally achieving better results than the standard GSGP. On the other hand, results show accentuated overfitting issues on more di"cult regression datasets.Vanneschi, LeonardoBatista, João Eduardo Silva PombinhoRUNDinis, Marta Teixeira Rumina2024-11-04T16:24:23Z2024-10-292024-10-29T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10362/174532TID:203796500enginfo:eu-repo/semantics/openAccessreponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiainstacron:RCAAP2025-01-13T01:41:10Zoai:run.unl.pt:10362/174532Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireinfo@rcaap.ptopendoar:https://opendoar.ac.uk/repository/71602025-05-28T19:12:49.414903Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiafalse
dc.title.none.fl_str_mv Parameter Control in Geometric Semantic Genetic Programming
title Parameter Control in Geometric Semantic Genetic Programming
spellingShingle Parameter Control in Geometric Semantic Genetic Programming
Dinis, Marta Teixeira Rumina
Genetic Programming
Geometric Semantic Genetic Programming
Genetic Operators
Mutation
Parameter Control
Domínio/Área Científica::Ciências Naturais::Ciências da Computação e da Informação
title_short Parameter Control in Geometric Semantic Genetic Programming
title_full Parameter Control in Geometric Semantic Genetic Programming
title_fullStr Parameter Control in Geometric Semantic Genetic Programming
title_full_unstemmed Parameter Control in Geometric Semantic Genetic Programming
title_sort Parameter Control in Geometric Semantic Genetic Programming
author Dinis, Marta Teixeira Rumina
author_facet Dinis, Marta Teixeira Rumina
author_role author
dc.contributor.none.fl_str_mv Vanneschi, Leonardo
Batista, João Eduardo Silva Pombinho
RUN
dc.contributor.author.fl_str_mv Dinis, Marta Teixeira Rumina
dc.subject.por.fl_str_mv Genetic Programming
Geometric Semantic Genetic Programming
Genetic Operators
Mutation
Parameter Control
Domínio/Área Científica::Ciências Naturais::Ciências da Computação e da Informação
topic Genetic Programming
Geometric Semantic Genetic Programming
Genetic Operators
Mutation
Parameter Control
Domínio/Área Científica::Ciências Naturais::Ciências da Computação e da Informação
description Dissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Data Science
publishDate 2024
dc.date.none.fl_str_mv 2024-11-04T16:24:23Z
2024-10-29
2024-10-29T00:00:00Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10362/174532
TID:203796500
url http://hdl.handle.net/10362/174532
identifier_str_mv TID:203796500
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron:RCAAP
instname_str FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron_str RCAAP
institution RCAAP
reponame_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
collection Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository.name.fl_str_mv Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
repository.mail.fl_str_mv info@rcaap.pt
_version_ 1833597945807634432

Similar Items