Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks

Detalhes bibliográficos
Autor(a) principal: Marar, João Fernando [UNESP]
Data de Publicação: 2008
Outros Autores: Coelho, Helder
Tipo de documento: Artigo de conferência
Idioma: eng
Título da fonte: Repositório Institucional da UNESP
Texto Completo: http://hdl.handle.net/11449/8307
Resumo: Wavelet functions have been used as the activation function in feedforward neural networks. An abundance of R&D has been produced on wavelet neural network area. Some successful algorithms and applications in wavelet neural network have been developed and reported in the literature. However, most of the aforementioned reports impose many restrictions in the classical backpropagation algorithm, such as low dimensionality, tensor product of wavelets, parameters initialization, and, in general, the output is one dimensional, etc. In order to remove some of these restrictions, a family of polynomial wavelets generated from powers of sigmoid functions is presented. We described how a multidimensional wavelet neural networks based on these functions can be constructed, trained and applied in pattern recognition tasks. As an example of application for the method proposed, it is studied the exclusive-or (XOR) problem.
id UNSP_f4b8b130eb3b107af4d9ce0dc6a65b74
oai_identifier_str oai:repositorio.unesp.br:11449/8307
network_acronym_str UNSP
network_name_str Repositório Institucional da UNESP
repository_id_str 2946
spelling Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networksartificial neural networkfunction approximationpolynomial powers of sigmoid (PPS)wavelets functionsPPS-Wavelet neural networksactivation functionsfeedforward networksWavelet functions have been used as the activation function in feedforward neural networks. An abundance of R&D has been produced on wavelet neural network area. Some successful algorithms and applications in wavelet neural network have been developed and reported in the literature. However, most of the aforementioned reports impose many restrictions in the classical backpropagation algorithm, such as low dimensionality, tensor product of wavelets, parameters initialization, and, in general, the output is one dimensional, etc. In order to remove some of these restrictions, a family of polynomial wavelets generated from powers of sigmoid functions is presented. We described how a multidimensional wavelet neural networks based on these functions can be constructed, trained and applied in pattern recognition tasks. As an example of application for the method proposed, it is studied the exclusive-or (XOR) problem.São Paulo State Univ, Fac Ciencias, Dept Comp, Adapt Syst & Computat Intelligence Lab, São Paulo, BrazilSão Paulo State Univ, Fac Ciencias, Dept Comp, Adapt Syst & Computat Intelligence Lab, São Paulo, BrazilInsticc-inst Syst Technologies Information Control & CommunicationUniversidade Estadual Paulista (Unesp)Marar, João Fernando [UNESP]Coelho, Helder2014-05-20T13:25:59Z2014-05-20T13:25:59Z2008-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObject261-268Biosignals 2008: Proceedings of The First International Conference on Bio-inspired Systems and Signal Processing, Vol Ii. Setubal: Insticc-inst Syst Technologies Information Control & Communication, p. 261-268, 2008.http://hdl.handle.net/11449/8307WOS:0002569831000441233049484488761Web of Sciencereponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengBiosignals 2008: Proceedings of The First International Conference on Bio-inspired Systems and Signal Processing, Vol Iiinfo:eu-repo/semantics/openAccess2024-04-23T16:11:27Zoai:repositorio.unesp.br:11449/8307Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T20:17:56.839254Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false
dc.title.none.fl_str_mv Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks
title Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks
spellingShingle Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks
Marar, João Fernando [UNESP]
artificial neural network
function approximation
polynomial powers of sigmoid (PPS)
wavelets functions
PPS-Wavelet neural networks
activation functions
feedforward networks
title_short Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks
title_full Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks
title_fullStr Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks
title_full_unstemmed Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks
title_sort Multidimensional polynomial powers of sigmoid (PPS) Wavelet neural networks
author Marar, João Fernando [UNESP]
author_facet Marar, João Fernando [UNESP]
Coelho, Helder
author_role author
author2 Coelho, Helder
author2_role author
dc.contributor.none.fl_str_mv Universidade Estadual Paulista (Unesp)
dc.contributor.author.fl_str_mv Marar, João Fernando [UNESP]
Coelho, Helder
dc.subject.por.fl_str_mv artificial neural network
function approximation
polynomial powers of sigmoid (PPS)
wavelets functions
PPS-Wavelet neural networks
activation functions
feedforward networks
topic artificial neural network
function approximation
polynomial powers of sigmoid (PPS)
wavelets functions
PPS-Wavelet neural networks
activation functions
feedforward networks
description Wavelet functions have been used as the activation function in feedforward neural networks. An abundance of R&D has been produced on wavelet neural network area. Some successful algorithms and applications in wavelet neural network have been developed and reported in the literature. However, most of the aforementioned reports impose many restrictions in the classical backpropagation algorithm, such as low dimensionality, tensor product of wavelets, parameters initialization, and, in general, the output is one dimensional, etc. In order to remove some of these restrictions, a family of polynomial wavelets generated from powers of sigmoid functions is presented. We described how a multidimensional wavelet neural networks based on these functions can be constructed, trained and applied in pattern recognition tasks. As an example of application for the method proposed, it is studied the exclusive-or (XOR) problem.
publishDate 2008
dc.date.none.fl_str_mv 2008-01-01
2014-05-20T13:25:59Z
2014-05-20T13:25:59Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/conferenceObject
format conferenceObject
status_str publishedVersion
dc.identifier.uri.fl_str_mv Biosignals 2008: Proceedings of The First International Conference on Bio-inspired Systems and Signal Processing, Vol Ii. Setubal: Insticc-inst Syst Technologies Information Control & Communication, p. 261-268, 2008.
http://hdl.handle.net/11449/8307
WOS:000256983100044
1233049484488761
identifier_str_mv Biosignals 2008: Proceedings of The First International Conference on Bio-inspired Systems and Signal Processing, Vol Ii. Setubal: Insticc-inst Syst Technologies Information Control & Communication, p. 261-268, 2008.
WOS:000256983100044
1233049484488761
url http://hdl.handle.net/11449/8307
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv Biosignals 2008: Proceedings of The First International Conference on Bio-inspired Systems and Signal Processing, Vol Ii
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv 261-268
dc.publisher.none.fl_str_mv Insticc-inst Syst Technologies Information Control & Communication
publisher.none.fl_str_mv Insticc-inst Syst Technologies Information Control & Communication
dc.source.none.fl_str_mv Web of Science
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_ 1808129185395245056