Regularization of extreme learning machines with information of spatial relations of the projected data

dc.creatorLourenço Ribeiro Grossi Araújo
dc.creatorLuiz C. B. Torres
dc.creatorCarla Caldeira Takahashi
dc.creatorAntônio de Pádua Braga
dc.creatorLeonardo J. Silvestre
dc.date.accessioned2025-05-13T15:08:57Z
dc.date.accessioned2025-09-08T22:54:32Z
dc.date.available2025-05-13T15:08:57Z
dc.date.issued2019
dc.identifier.doi10.1109/CoDIT.2019.8820610
dc.identifier.urihttps://hdl.handle.net/1843/82230
dc.languageeng
dc.publisherUniversidade Federal de Minas Gerais
dc.relation.ispartof6th International Conference on Control, Decision and Information Technologies (CoDIT)
dc.rightsAcesso Restrito
dc.subjectRedes neurais (Computação)
dc.subject.otherNeurons , Training , Cybernetics , Minimization , Testing , Biological neural networks , Aerospace electronics
dc.subject.otherExtreme Learning Machine , Training Time , Parameter Selection , Weight Decay , Regularization Parameter , Null Hypothesis , High-dimensional , Dimensional Space , Hidden Layer , Weight Matrix , Linear Discriminant Analysis , Weight Vector , Decision Boundary , Neurons In The Hidden Layer , Hidden Neurons , Affinity Matrix , Linearly Separable , Regularization Scheme , Class Mean , Universal Approximation , Norm Penalty , Fisher Score , Single Layer Feedforward Network , Empirical Risk Minimization
dc.titleRegularization of extreme learning machines with information of spatial relations of the projected data
dc.typeArtigo de evento
local.description.resumoThe following work presents a new approach to automatic selection of Tikhonov's regularization parameter, responsible for controlling the weight value of an ELM neural network. A strategy based on measurements obtained from data projection (Fisher-Score) is introduced. Seven datasets are tested and results are compared to those obtained when the regularization parameter is selected through cross-validation. The strategy shows satisfactory classification performance (in terms of p-value), while presenting significant training time reduction.
local.publisher.countryBrasil
local.publisher.departmentENG - DEPARTAMENTO DE ENGENHARIA ELETRÔNICA
local.publisher.initialsUFMG
local.url.externahttps://ieeexplore.ieee.org/document/8820610

Arquivos

Licença do pacote

Agora exibindo 1 - 1 de 1
Carregando...
Imagem de Miniatura
Nome:
License.txt
Tamanho:
1.99 KB
Formato:
Plain Text
Descrição: