Fine-Tuning Temperatures in Restricted Boltzmann Machines Using Meta-Heuristic Optimization

Registro completo de metadados
MetadadosDescriçãoIdioma
Autor(es): dc.contributorUniversidade Estadual Paulista (UNESP)-
Autor(es): dc.creatorRoder, Mateus [UNESP]-
Autor(es): dc.creatorRosa, Gustavo Henrique de [UNESP]-
Autor(es): dc.creatorPapa, Joao Paulo [UNESP]-
Autor(es): dc.creatorBreve, Fabricio Aparecido [UNESP]-
Autor(es): dc.creatorIEEE-
Data de aceite: dc.date.accessioned2022-08-04T21:58:53Z-
Data de disponibilização: dc.date.available2022-08-04T21:58:53Z-
Data de envio: dc.date.issued2022-04-28-
Data de envio: dc.date.issued2022-04-28-
Data de envio: dc.date.issued2019-12-31-
Fonte completa do material: dc.identifierhttp://hdl.handle.net/11449/218678-
Fonte: dc.identifier.urihttp://educapes.capes.gov.br/handle/11449/218678-
Descrição: dc.descriptionRestricted Boltzmann Machines (RBM) are stochastic neural networks mainly used for image reconstruction and unsupervised feature learning. An enhanced version, the temperature-based RBM (T-RBM), considers a new temperature parameter during the learning process that influences the neurons' activation. Nevertheless, the major vulnerability of such models concerns selecting an adequate system's temperature, which might lead them to inadequate training or even overfitting when wrongly set, thus limiting the network from predicting or working effectively over unseen data. This paper addresses the problem of selecting a suitable system's temperature through a meta-heuristic optimization process. Meta-heuristic-driven techniques, such as Particle Swarm Optimization, Bat Algorithm, and Artificial Bee Colony are employed to find proper values for the temperature parameter. Additionally, for comparison purposes, three standard temperature values and a random search are used as baselines. The results revealed that optimizing T-RBM is suitable for training purposes, primarily due to their complex fitness landscape, which makes fine-tuning temperatures a non-trivial task.-
Descrição: dc.descriptionFundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)-
Descrição: dc.descriptionConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)-
Descrição: dc.descriptionUNESP Sao Paulo State Univ, Dept Comp, Bauru, SP, Brazil-
Descrição: dc.descriptionSao Paulo State Univ, Dept Math Stat & Comp, Rio Claro, SP, Brazil-
Descrição: dc.descriptionUNESP Sao Paulo State Univ, Dept Comp, Bauru, SP, Brazil-
Descrição: dc.descriptionSao Paulo State Univ, Dept Math Stat & Comp, Rio Claro, SP, Brazil-
Descrição: dc.descriptionFAPESP: 2013/07375-0-
Descrição: dc.descriptionFAPESP: 2014/12236-1-
Descrição: dc.descriptionFAPESP: 2017/25908-6-
Descrição: dc.descriptionFAPESP: 2019/02205-5-
Descrição: dc.descriptionFAPESP: 2019/07825-1-
Descrição: dc.descriptionCNPq: 307066/2017-7-
Descrição: dc.descriptionCNPq: 427968/2018-6-
Formato: dc.format8-
Idioma: dc.languageen-
Publicador: dc.publisherIeee-
Relação: dc.relation2020 Ieee Congress On Evolutionary Computation (cec)-
???dc.source???: dc.sourceWeb of Science-
Palavras-chave: dc.subjectImage Reconstruction-
Palavras-chave: dc.subjectRestricted Boltzmann Machine-
Palavras-chave: dc.subjectTemperature-based Systems-
Palavras-chave: dc.subjectMeta-Heuristic Optimization-
Título: dc.titleFine-Tuning Temperatures in Restricted Boltzmann Machines Using Meta-Heuristic Optimization-
Aparece nas coleções:Repositório Institucional - Unesp

Não existem arquivos associados a este item.