Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition

Autores
Aguirre, Fernando Leonel; Gomez, Nicolás M.; Pazos, Sebastián Matías; Palumbo, Félix Roberto Mario; Suñé, Jordi; Miranda, Enrique
Año de publicación
2021
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
In this paper, we extend the application of the Quasi-Static Memdiode model to the real-istic SPICE simulation of memristor-based single (SLPs) and multilayer perceptrons (MLPs) in-tended for large dataset pattern recognition. By considering ex-situ training and the classification of the hand-written characters of the MNIST database, we evaluate the degradation of the inference accuracy due to the interconnection resistances for MLPs involving up to three hidden neural layers. Two approaches to reduce the impact of the line resistance are considered and implemented in our simulations, they are the inclusion of an iterative calibration algorithm and the partitioning of the synaptic layers into smaller blocks. The obtained results indicate that MLPs are more sensitive to the line resistance effect than SLPs and that partitioning is the most effective way to minimize the impact of high line resistance values.
Fil: Aguirre, Fernando Leonel. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Gomez, Nicolás M.. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; Argentina
Fil: Pazos, Sebastián Matías. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; Argentina
Fil: Palumbo, Félix Roberto Mario. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; Argentina
Fil: Suñé, Jordi. Universitat Autònoma de Barcelona; España
Fil: Miranda, Enrique. Universitat Autònoma de Barcelona; España
Materia
CROSS-POINT
MEMORY
MEMRISTOR
MULTILAYER PERCEPTRON
NEUROMORPHIC
PATTERN RECOGNITION
RESISTIVE-SWITCHING
RRAM
Nivel de accesibilidad
acceso abierto
Condiciones de uso
https://creativecommons.org/licenses/by/2.5/ar/
Repositorio
CONICET Digital (CONICET)
Institución
Consejo Nacional de Investigaciones Científicas y Técnicas
OAI Identificador
oai:ri.conicet.gov.ar:11336/165149
Acceder
id CONICETDig_36e8a257a2fd05a9fee4a77305e42c80
oai_identifier_str oai:ri.conicet.gov.ar:11336/165149
network_acronym_str CONICETDig
repository_id_str 3498
network_name_str CONICET Digital (CONICET)
spelling Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognitionAguirre, Fernando LeonelGomez, Nicolás M.Pazos, Sebastián MatíasPalumbo, Félix Roberto MarioSuñé, JordiMiranda, EnriqueCROSS-POINTMEMORYMEMRISTORMULTILAYER PERCEPTRONNEUROMORPHICPATTERN RECOGNITIONRESISTIVE-SWITCHINGRRAMhttps://purl.org/becyt/ford/1.3https://purl.org/becyt/ford/1In this paper, we extend the application of the Quasi-Static Memdiode model to the real-istic SPICE simulation of memristor-based single (SLPs) and multilayer perceptrons (MLPs) in-tended for large dataset pattern recognition. By considering ex-situ training and the classification of the hand-written characters of the MNIST database, we evaluate the degradation of the inference accuracy due to the interconnection resistances for MLPs involving up to three hidden neural layers. Two approaches to reduce the impact of the line resistance are considered and implemented in our simulations, they are the inclusion of an iterative calibration algorithm and the partitioning of the synaptic layers into smaller blocks. The obtained results indicate that MLPs are more sensitive to the line resistance effect than SLPs and that partitioning is the most effective way to minimize the impact of high line resistance values.Fil: Aguirre, Fernando Leonel. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; ArgentinaFil: Gomez, Nicolás M.. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; ArgentinaFil: Pazos, Sebastián Matías. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; ArgentinaFil: Palumbo, Félix Roberto Mario. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; ArgentinaFil: Suñé, Jordi. Universitat Autònoma de Barcelona; EspañaFil: Miranda, Enrique. Universitat Autònoma de Barcelona; EspañaMDPI AG2021-03info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articuloapplication/pdfapplication/pdfapplication/pdfapplication/pdfhttp://hdl.handle.net/11336/165149Aguirre, Fernando Leonel; Gomez, Nicolás M.; Pazos, Sebastián Matías; Palumbo, Félix Roberto Mario; Suñé, Jordi; et al.; Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition; MDPI AG; Journal of Low Power Electronics and Applications; 11; 1; 3-2021; 1-182079-9268CONICET DigitalCONICETenginfo:eu-repo/semantics/altIdentifier/url/https://www.mdpi.com/2079-9268/11/1/9info:eu-repo/semantics/altIdentifier/doi/10.3390/jlpea11010009info:eu-repo/semantics/openAccesshttps://creativecommons.org/licenses/by/2.5/ar/reponame:CONICET Digital (CONICET)instname:Consejo Nacional de Investigaciones Científicas y Técnicas2025-09-10T13:21:08Zoai:ri.conicet.gov.ar:11336/165149instacron:CONICETInstitucionalhttp://ri.conicet.gov.ar/Organismo científico-tecnológicoNo correspondehttp://ri.conicet.gov.ar/oai/requestdasensio@conicet.gov.ar; lcarlino@conicet.gov.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:34982025-09-10 13:21:08.383CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicasfalse
dc.title.none.fl_str_mv Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition
title Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition
spellingShingle Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition
Aguirre, Fernando Leonel
CROSS-POINT
MEMORY
MEMRISTOR
MULTILAYER PERCEPTRON
NEUROMORPHIC
PATTERN RECOGNITION
RESISTIVE-SWITCHING
RRAM
title_short Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition
title_full Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition
title_fullStr Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition
title_full_unstemmed Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition
title_sort Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition
dc.creator.none.fl_str_mv Aguirre, Fernando Leonel
Gomez, Nicolás M.
Pazos, Sebastián Matías
Palumbo, Félix Roberto Mario
Suñé, Jordi
Miranda, Enrique
author Aguirre, Fernando Leonel
author_facet Aguirre, Fernando Leonel
Gomez, Nicolás M.
Pazos, Sebastián Matías
Palumbo, Félix Roberto Mario
Suñé, Jordi
Miranda, Enrique
author_role author
author2 Gomez, Nicolás M.
Pazos, Sebastián Matías
Palumbo, Félix Roberto Mario
Suñé, Jordi
Miranda, Enrique
author2_role author
author
author
author
author
dc.subject.none.fl_str_mv CROSS-POINT
MEMORY
MEMRISTOR
MULTILAYER PERCEPTRON
NEUROMORPHIC
PATTERN RECOGNITION
RESISTIVE-SWITCHING
RRAM
topic CROSS-POINT
MEMORY
MEMRISTOR
MULTILAYER PERCEPTRON
NEUROMORPHIC
PATTERN RECOGNITION
RESISTIVE-SWITCHING
RRAM
purl_subject.fl_str_mv https://purl.org/becyt/ford/1.3
https://purl.org/becyt/ford/1
dc.description.none.fl_txt_mv In this paper, we extend the application of the Quasi-Static Memdiode model to the real-istic SPICE simulation of memristor-based single (SLPs) and multilayer perceptrons (MLPs) in-tended for large dataset pattern recognition. By considering ex-situ training and the classification of the hand-written characters of the MNIST database, we evaluate the degradation of the inference accuracy due to the interconnection resistances for MLPs involving up to three hidden neural layers. Two approaches to reduce the impact of the line resistance are considered and implemented in our simulations, they are the inclusion of an iterative calibration algorithm and the partitioning of the synaptic layers into smaller blocks. The obtained results indicate that MLPs are more sensitive to the line resistance effect than SLPs and that partitioning is the most effective way to minimize the impact of high line resistance values.
Fil: Aguirre, Fernando Leonel. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Gomez, Nicolás M.. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; Argentina
Fil: Pazos, Sebastián Matías. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; Argentina
Fil: Palumbo, Félix Roberto Mario. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina. Universidad Tecnológica Nacional. Facultad Regional Buenos Aires; Argentina
Fil: Suñé, Jordi. Universitat Autònoma de Barcelona; España
Fil: Miranda, Enrique. Universitat Autònoma de Barcelona; España
description In this paper, we extend the application of the Quasi-Static Memdiode model to the real-istic SPICE simulation of memristor-based single (SLPs) and multilayer perceptrons (MLPs) in-tended for large dataset pattern recognition. By considering ex-situ training and the classification of the hand-written characters of the MNIST database, we evaluate the degradation of the inference accuracy due to the interconnection resistances for MLPs involving up to three hidden neural layers. Two approaches to reduce the impact of the line resistance are considered and implemented in our simulations, they are the inclusion of an iterative calibration algorithm and the partitioning of the synaptic layers into smaller blocks. The obtained results indicate that MLPs are more sensitive to the line resistance effect than SLPs and that partitioning is the most effective way to minimize the impact of high line resistance values.
publishDate 2021
dc.date.none.fl_str_mv 2021-03
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/11336/165149
Aguirre, Fernando Leonel; Gomez, Nicolás M.; Pazos, Sebastián Matías; Palumbo, Félix Roberto Mario; Suñé, Jordi; et al.; Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition; MDPI AG; Journal of Low Power Electronics and Applications; 11; 1; 3-2021; 1-18
2079-9268
CONICET Digital
CONICET
url http://hdl.handle.net/11336/165149
identifier_str_mv Aguirre, Fernando Leonel; Gomez, Nicolás M.; Pazos, Sebastián Matías; Palumbo, Félix Roberto Mario; Suñé, Jordi; et al.; Minimization of the line resistance impact on memdiode-based simulations of multilayer perceptron arrays applied to pattern recognition; MDPI AG; Journal of Low Power Electronics and Applications; 11; 1; 3-2021; 1-18
2079-9268
CONICET Digital
CONICET
dc.language.none.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv info:eu-repo/semantics/altIdentifier/url/https://www.mdpi.com/2079-9268/11/1/9
info:eu-repo/semantics/altIdentifier/doi/10.3390/jlpea11010009
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
https://creativecommons.org/licenses/by/2.5/ar/
eu_rights_str_mv openAccess
rights_invalid_str_mv https://creativecommons.org/licenses/by/2.5/ar/
dc.format.none.fl_str_mv application/pdf
application/pdf
application/pdf
application/pdf
dc.publisher.none.fl_str_mv MDPI AG
publisher.none.fl_str_mv MDPI AG
dc.source.none.fl_str_mv reponame:CONICET Digital (CONICET)
instname:Consejo Nacional de Investigaciones Científicas y Técnicas
reponame_str CONICET Digital (CONICET)
collection CONICET Digital (CONICET)
instname_str Consejo Nacional de Investigaciones Científicas y Técnicas
repository.name.fl_str_mv CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicas
repository.mail.fl_str_mv dasensio@conicet.gov.ar; lcarlino@conicet.gov.ar
_version_ 1842981158427557888
score 12.48226

Publicaciones similares