Convolutional neural networks ensembles through single-iteration optimization
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1007/s00500-022-06791-9 http://hdl.handle.net/11449/234078 |
Resumo: | Convolutional Neural Networks have been widely employed in a diverse range of computer vision-based applications, including image classification, object recognition, and object segmentation. Nevertheless, one weakness of such models concerns their hyperparameters’ setting, being highly specific for each particular problem. One common approach is to employ meta-heuristic optimization algorithms to find suitable sets of hyperparameters at the expense of increasing the computational burden, being unfeasible under real-time scenarios. In this paper, we address this problem by creating Convolutional Neural Networks ensembles through Single-Iteration Optimization, a fast optimization composed of only one iteration that is no more effective than a random search. Essentially, the idea is to provide the same capability offered by long-term optimizations, however, without their computational loads. The results among four well-known datasets revealed that creating one-iteration optimized ensembles provides promising results while diminishing the time to achieve them. |
id |
UNSP_e0f49982d354eb0b954f583a78528faf |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/234078 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Convolutional neural networks ensembles through single-iteration optimizationConvolutional neural networksEnsemblesMachine learningMeta-heuristicsOptimizationConvolutional Neural Networks have been widely employed in a diverse range of computer vision-based applications, including image classification, object recognition, and object segmentation. Nevertheless, one weakness of such models concerns their hyperparameters’ setting, being highly specific for each particular problem. One common approach is to employ meta-heuristic optimization algorithms to find suitable sets of hyperparameters at the expense of increasing the computational burden, being unfeasible under real-time scenarios. In this paper, we address this problem by creating Convolutional Neural Networks ensembles through Single-Iteration Optimization, a fast optimization composed of only one iteration that is no more effective than a random search. Essentially, the idea is to provide the same capability offered by long-term optimizations, however, without their computational loads. The results among four well-known datasets revealed that creating one-iteration optimized ensembles provides promising results while diminishing the time to achieve them.Department of Computing São Paulo State UniversityDepartment of Computing São Paulo State UniversityUniversidade Estadual Paulista (UNESP)Ribeiro, Luiz C. F. [UNESP]Rosa, Gustavo H. de [UNESP]Rodrigues, Douglas [UNESP]Papa, João P. [UNESP]2022-05-01T13:11:37Z2022-05-01T13:11:37Z2022-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://dx.doi.org/10.1007/s00500-022-06791-9Soft Computing.1433-74791432-7643http://hdl.handle.net/11449/23407810.1007/s00500-022-06791-92-s2.0-85123853757Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengSoft Computinginfo:eu-repo/semantics/openAccess2024-04-23T16:10:49Zoai:repositorio.unesp.br:11449/234078Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T18:38:39.814258Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Convolutional neural networks ensembles through single-iteration optimization |
title |
Convolutional neural networks ensembles through single-iteration optimization |
spellingShingle |
Convolutional neural networks ensembles through single-iteration optimization Ribeiro, Luiz C. F. [UNESP] Convolutional neural networks Ensembles Machine learning Meta-heuristics Optimization |
title_short |
Convolutional neural networks ensembles through single-iteration optimization |
title_full |
Convolutional neural networks ensembles through single-iteration optimization |
title_fullStr |
Convolutional neural networks ensembles through single-iteration optimization |
title_full_unstemmed |
Convolutional neural networks ensembles through single-iteration optimization |
title_sort |
Convolutional neural networks ensembles through single-iteration optimization |
author |
Ribeiro, Luiz C. F. [UNESP] |
author_facet |
Ribeiro, Luiz C. F. [UNESP] Rosa, Gustavo H. de [UNESP] Rodrigues, Douglas [UNESP] Papa, João P. [UNESP] |
author_role |
author |
author2 |
Rosa, Gustavo H. de [UNESP] Rodrigues, Douglas [UNESP] Papa, João P. [UNESP] |
author2_role |
author author author |
dc.contributor.none.fl_str_mv |
Universidade Estadual Paulista (UNESP) |
dc.contributor.author.fl_str_mv |
Ribeiro, Luiz C. F. [UNESP] Rosa, Gustavo H. de [UNESP] Rodrigues, Douglas [UNESP] Papa, João P. [UNESP] |
dc.subject.por.fl_str_mv |
Convolutional neural networks Ensembles Machine learning Meta-heuristics Optimization |
topic |
Convolutional neural networks Ensembles Machine learning Meta-heuristics Optimization |
description |
Convolutional Neural Networks have been widely employed in a diverse range of computer vision-based applications, including image classification, object recognition, and object segmentation. Nevertheless, one weakness of such models concerns their hyperparameters’ setting, being highly specific for each particular problem. One common approach is to employ meta-heuristic optimization algorithms to find suitable sets of hyperparameters at the expense of increasing the computational burden, being unfeasible under real-time scenarios. In this paper, we address this problem by creating Convolutional Neural Networks ensembles through Single-Iteration Optimization, a fast optimization composed of only one iteration that is no more effective than a random search. Essentially, the idea is to provide the same capability offered by long-term optimizations, however, without their computational loads. The results among four well-known datasets revealed that creating one-iteration optimized ensembles provides promising results while diminishing the time to achieve them. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-05-01T13:11:37Z 2022-05-01T13:11:37Z 2022-01-01 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1007/s00500-022-06791-9 Soft Computing. 1433-7479 1432-7643 http://hdl.handle.net/11449/234078 10.1007/s00500-022-06791-9 2-s2.0-85123853757 |
url |
http://dx.doi.org/10.1007/s00500-022-06791-9 http://hdl.handle.net/11449/234078 |
identifier_str_mv |
Soft Computing. 1433-7479 1432-7643 10.1007/s00500-022-06791-9 2-s2.0-85123853757 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Soft Computing |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1808128958966792192 |