Liquid State Machine on SpiNNaker for Spatio-Temporal Classification Tasks

Liquid State Machines (LSMs) are computing reservoirs composed of recurrently connected Spiking Neural Networks which have attracted research interest for their modeling capacity of biological structures and as promising pattern recognition tools suitable for their implementation in neuromorphic pro...

Descripción completa

Detalles Bibliográficos
Autores: Patiño Saucedo, Alberto, Rostro González, Horacio, Serrano Gotarredona, María Teresa, Linares Barranco, Bernabé
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2022
País:España
Institución:Universidad de Sevilla (US)
Repositorio:idUS. Depósito de Investigación de la Universidad de Sevilla
OAI Identifier:oai:idus.us.es:11441/144004
Acceso en línea:https://hdl.handle.net/11441/144004
https://doi.org/10.3389/fnins.2022.819063
Access Level:acceso abierto
Palabra clave:Liquid State Machine
N-MNIST
neuromorphic hardware
spiking neural network
SpiNNaker
id ES_fc4e64e20ee0fa7a19ee4e274982f81a
oai_identifier_str oai:idus.us.es:11441/144004
network_acronym_str ES
network_name_str España
repository_id_str
spelling Liquid State Machine on SpiNNaker for Spatio-Temporal Classification TasksPatiño Saucedo, AlbertoRostro González, HoracioSerrano Gotarredona, María TeresaLinares Barranco, BernabéLiquid State MachineN-MNISTneuromorphic hardwarespiking neural networkSpiNNakerLiquid State Machines (LSMs) are computing reservoirs composed of recurrently connected Spiking Neural Networks which have attracted research interest for their modeling capacity of biological structures and as promising pattern recognition tools suitable for their implementation in neuromorphic processors, benefited from the modest use of computing resources in their training process. However, it has been difficult to optimize LSMs for solving complex tasks such as event-based computer vision and few implementations in large-scale neuromorphic processors have been attempted. In this work, we show that offline-trained LSMs implemented in the SpiNNaker neuromorphic processor are able to classify visual events, achieving state-of-the-art performance in the event-based N-MNIST dataset. The training of the readout layer is performed using a recent adaptation of back-propagation-through-time (BPTT) for SNNs, while the internal weights of the reservoir are kept static. Results show that mapping our LSM from a Deep Learning framework to SpiNNaker does not affect the performance of the classification task. Additionally, we show that weight quantization, which substantially reduces the memory footprint of the LSM, has a small impact on its performance.Frontiers Media S.A.Arquitectura y Tecnología de Computadores2022info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttps://hdl.handle.net/11441/144004https://doi.org/10.3389/fnins.2022.819063reponame:idUS. Depósito de Investigación de la Universidad de Sevillainstname:Universidad de Sevilla (US)InglésFrontiers in Neuroscience, 16.https://www.frontiersin.org/articles/10.3389/fnins.2022.819063/fullinfo:eu-repo/semantics/openAccessoai:idus.us.es:11441/1440042026-06-17T12:51:07Z
dc.title.none.fl_str_mv Liquid State Machine on SpiNNaker for Spatio-Temporal Classification Tasks
title Liquid State Machine on SpiNNaker for Spatio-Temporal Classification Tasks
spellingShingle Liquid State Machine on SpiNNaker for Spatio-Temporal Classification Tasks
Patiño Saucedo, Alberto
Liquid State Machine
N-MNIST
neuromorphic hardware
spiking neural network
SpiNNaker
title_short Liquid State Machine on SpiNNaker for Spatio-Temporal Classification Tasks
title_full Liquid State Machine on SpiNNaker for Spatio-Temporal Classification Tasks
title_fullStr Liquid State Machine on SpiNNaker for Spatio-Temporal Classification Tasks
title_full_unstemmed Liquid State Machine on SpiNNaker for Spatio-Temporal Classification Tasks
title_sort Liquid State Machine on SpiNNaker for Spatio-Temporal Classification Tasks
dc.creator.none.fl_str_mv Patiño Saucedo, Alberto
Rostro González, Horacio
Serrano Gotarredona, María Teresa
Linares Barranco, Bernabé
author Patiño Saucedo, Alberto
author_facet Patiño Saucedo, Alberto
Rostro González, Horacio
Serrano Gotarredona, María Teresa
Linares Barranco, Bernabé
author_role author
author2 Rostro González, Horacio
Serrano Gotarredona, María Teresa
Linares Barranco, Bernabé
author2_role author
author
author
dc.contributor.none.fl_str_mv Arquitectura y Tecnología de Computadores
dc.subject.none.fl_str_mv Liquid State Machine
N-MNIST
neuromorphic hardware
spiking neural network
SpiNNaker
topic Liquid State Machine
N-MNIST
neuromorphic hardware
spiking neural network
SpiNNaker
description Liquid State Machines (LSMs) are computing reservoirs composed of recurrently connected Spiking Neural Networks which have attracted research interest for their modeling capacity of biological structures and as promising pattern recognition tools suitable for their implementation in neuromorphic processors, benefited from the modest use of computing resources in their training process. However, it has been difficult to optimize LSMs for solving complex tasks such as event-based computer vision and few implementations in large-scale neuromorphic processors have been attempted. In this work, we show that offline-trained LSMs implemented in the SpiNNaker neuromorphic processor are able to classify visual events, achieving state-of-the-art performance in the event-based N-MNIST dataset. The training of the readout layer is performed using a recent adaptation of back-propagation-through-time (BPTT) for SNNs, while the internal weights of the reservoir are kept static. Results show that mapping our LSM from a Deep Learning framework to SpiNNaker does not affect the performance of the classification task. Additionally, we show that weight quantization, which substantially reduces the memory footprint of the LSM, has a small impact on its performance.
publishDate 2022
dc.date.none.fl_str_mv 2022
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv https://hdl.handle.net/11441/144004
https://doi.org/10.3389/fnins.2022.819063
url https://hdl.handle.net/11441/144004
https://doi.org/10.3389/fnins.2022.819063
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Frontiers in Neuroscience, 16.
https://www.frontiersin.org/articles/10.3389/fnins.2022.819063/full
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv Frontiers Media S.A.
publisher.none.fl_str_mv Frontiers Media S.A.
dc.source.none.fl_str_mv reponame:idUS. Depósito de Investigación de la Universidad de Sevilla
instname:Universidad de Sevilla (US)
instname_str Universidad de Sevilla (US)
reponame_str idUS. Depósito de Investigación de la Universidad de Sevilla
collection idUS. Depósito de Investigación de la Universidad de Sevilla
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869425410146566144
score 15,300724

Similares en LA Referencia