Neural networks with high-order connections
Autor(a) principal: | |
---|---|
Data de Publicação: | 1993 |
Outros Autores: | |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFRGS |
Texto Completo: | http://hdl.handle.net/10183/101306 |
Resumo: | We present results for two difFerent kinds of high-order connections between neurons acting as corrections to the Hopfield model. Equilibrium properties are analyzed using the replica mean-field theory and compared with numerical simulations. An optimal learning algorithm for fourth-order connections is given that improves the storage capacity without increasing the weight of the higherorder term. While the behavior of one of the models qualitatively resembles the original Hopfield one, the other presents a new and very rich behavior: depending on the strength of the fourth-order connections and the temperature, the system presents two distinct retrieval regions separated by a gap, as well as several phase transitions. Also, the spin-glass states seems to disappear above a certain value of the load parameter α, αg. |
id |
UFRGS-2_6d41c44de408119dcb25c9ed301c000b |
---|---|
oai_identifier_str |
oai:www.lume.ufrgs.br:10183/101306 |
network_acronym_str |
UFRGS-2 |
network_name_str |
Repositório Institucional da UFRGS |
repository_id_str |
|
spelling |
Arenzon, Jeferson JacobAlmeida, Rita Maria Cunha de2014-08-19T02:10:10Z19931063-651Xhttp://hdl.handle.net/10183/101306000056962We present results for two difFerent kinds of high-order connections between neurons acting as corrections to the Hopfield model. Equilibrium properties are analyzed using the replica mean-field theory and compared with numerical simulations. An optimal learning algorithm for fourth-order connections is given that improves the storage capacity without increasing the weight of the higherorder term. While the behavior of one of the models qualitatively resembles the original Hopfield one, the other presents a new and very rich behavior: depending on the strength of the fourth-order connections and the temperature, the system presents two distinct retrieval regions separated by a gap, as well as several phase transitions. Also, the spin-glass states seems to disappear above a certain value of the load parameter α, αg.application/pdfengPhysical Review. E, Statistical Physics, Plasmas, Fluids and Related Interdisciplinary Topics. New York. Vol. 48, no. 5 (Nov. 1993), p. 4060-4069Física da matéria condensadaRedes neuraisBiofísicaModelos de cerebroNeural networks with high-order connectionsEstrangeiroinfo:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/openAccessreponame:Repositório Institucional da UFRGSinstname:Universidade Federal do Rio Grande do Sul (UFRGS)instacron:UFRGSORIGINAL000056962.pdf000056962.pdfTexto completo (inglês)application/pdf534056http://www.lume.ufrgs.br/bitstream/10183/101306/1/000056962.pdfe6299209fe97f5eca6f80d8aea53ed6dMD51TEXT000056962.pdf.txt000056962.pdf.txtExtracted Texttext/plain41291http://www.lume.ufrgs.br/bitstream/10183/101306/2/000056962.pdf.txtf0ff359d5530546b9068749a2abdfca8MD52THUMBNAIL000056962.pdf.jpg000056962.pdf.jpgGenerated Thumbnailimage/jpeg1911http://www.lume.ufrgs.br/bitstream/10183/101306/3/000056962.pdf.jpg74180971338e433ec2299c4511cf91eeMD5310183/1013062024-03-29 06:19:36.086078oai:www.lume.ufrgs.br:10183/101306Repositório de PublicaçõesPUBhttps://lume.ufrgs.br/oai/requestopendoar:2024-03-29T09:19:36Repositório Institucional da UFRGS - Universidade Federal do Rio Grande do Sul (UFRGS)false |
dc.title.pt_BR.fl_str_mv |
Neural networks with high-order connections |
title |
Neural networks with high-order connections |
spellingShingle |
Neural networks with high-order connections Arenzon, Jeferson Jacob Física da matéria condensada Redes neurais Biofísica Modelos de cerebro |
title_short |
Neural networks with high-order connections |
title_full |
Neural networks with high-order connections |
title_fullStr |
Neural networks with high-order connections |
title_full_unstemmed |
Neural networks with high-order connections |
title_sort |
Neural networks with high-order connections |
author |
Arenzon, Jeferson Jacob |
author_facet |
Arenzon, Jeferson Jacob Almeida, Rita Maria Cunha de |
author_role |
author |
author2 |
Almeida, Rita Maria Cunha de |
author2_role |
author |
dc.contributor.author.fl_str_mv |
Arenzon, Jeferson Jacob Almeida, Rita Maria Cunha de |
dc.subject.por.fl_str_mv |
Física da matéria condensada Redes neurais Biofísica Modelos de cerebro |
topic |
Física da matéria condensada Redes neurais Biofísica Modelos de cerebro |
description |
We present results for two difFerent kinds of high-order connections between neurons acting as corrections to the Hopfield model. Equilibrium properties are analyzed using the replica mean-field theory and compared with numerical simulations. An optimal learning algorithm for fourth-order connections is given that improves the storage capacity without increasing the weight of the higherorder term. While the behavior of one of the models qualitatively resembles the original Hopfield one, the other presents a new and very rich behavior: depending on the strength of the fourth-order connections and the temperature, the system presents two distinct retrieval regions separated by a gap, as well as several phase transitions. Also, the spin-glass states seems to disappear above a certain value of the load parameter α, αg. |
publishDate |
1993 |
dc.date.issued.fl_str_mv |
1993 |
dc.date.accessioned.fl_str_mv |
2014-08-19T02:10:10Z |
dc.type.driver.fl_str_mv |
Estrangeiro info:eu-repo/semantics/article |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10183/101306 |
dc.identifier.issn.pt_BR.fl_str_mv |
1063-651X |
dc.identifier.nrb.pt_BR.fl_str_mv |
000056962 |
identifier_str_mv |
1063-651X 000056962 |
url |
http://hdl.handle.net/10183/101306 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.ispartof.pt_BR.fl_str_mv |
Physical Review. E, Statistical Physics, Plasmas, Fluids and Related Interdisciplinary Topics. New York. Vol. 48, no. 5 (Nov. 1993), p. 4060-4069 |
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:Repositório Institucional da UFRGS instname:Universidade Federal do Rio Grande do Sul (UFRGS) instacron:UFRGS |
instname_str |
Universidade Federal do Rio Grande do Sul (UFRGS) |
instacron_str |
UFRGS |
institution |
UFRGS |
reponame_str |
Repositório Institucional da UFRGS |
collection |
Repositório Institucional da UFRGS |
bitstream.url.fl_str_mv |
http://www.lume.ufrgs.br/bitstream/10183/101306/1/000056962.pdf http://www.lume.ufrgs.br/bitstream/10183/101306/2/000056962.pdf.txt http://www.lume.ufrgs.br/bitstream/10183/101306/3/000056962.pdf.jpg |
bitstream.checksum.fl_str_mv |
e6299209fe97f5eca6f80d8aea53ed6d f0ff359d5530546b9068749a2abdfca8 74180971338e433ec2299c4511cf91ee |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFRGS - Universidade Federal do Rio Grande do Sul (UFRGS) |
repository.mail.fl_str_mv |
|
_version_ |
1801224845277528064 |