Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity

Abstract Spike timing-dependent plasticity (STDP), which is widely studied as a fundamental synaptic update rule for neuromorphic hardware, requires precise control of continuous weights. From the viewpoint of hardware implementation, a simplified update rule is desirable. Although simplified STDP w...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yoshifumi Nishi, Kumiko Nomura, Takao Marukame, Koichi Mizushima
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/1b42b9540ff943b4805ab44b662f914c
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:1b42b9540ff943b4805ab44b662f914c
record_format	dspace
spelling	oai:doaj.org-article:1b42b9540ff943b4805ab44b662f914c2021-12-02T18:02:14ZStochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity10.1038/s41598-021-97583-y2045-2322https://doaj.org/article/1b42b9540ff943b4805ab44b662f914c2021-09-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-97583-yhttps://doaj.org/toc/2045-2322Abstract Spike timing-dependent plasticity (STDP), which is widely studied as a fundamental synaptic update rule for neuromorphic hardware, requires precise control of continuous weights. From the viewpoint of hardware implementation, a simplified update rule is desirable. Although simplified STDP with stochastic binary synapses was proposed previously, we find that it leads to degradation of memory maintenance during learning, which is unfavourable for unsupervised online learning. In this work, we propose a stochastic binary synaptic model where the cumulative probability of the weight change evolves in a sigmoidal fashion with potentiation or depression trials, which can be implemented using a pair of switching devices consisting of serially connected multiple binary memristors. As a benchmark test we perform simulations of unsupervised learning of MNIST images with a two-layer network and show that simplified STDP in combination with this model can outperform conventional rules with continuous weights not only in memory maintenance but also in recognition accuracy. Our method achieves 97.3% in recognition accuracy, which is higher than that reported with standard STDP in the same framework. We also show that the high performance of our learning rule is robust against device-to-device variability of the memristor's probabilistic behaviour.Yoshifumi NishiKumiko NomuraTakao MarukameKoichi MizushimaNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-12 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Yoshifumi Nishi Kumiko Nomura Takao Marukame Koichi Mizushima Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity
description	Abstract Spike timing-dependent plasticity (STDP), which is widely studied as a fundamental synaptic update rule for neuromorphic hardware, requires precise control of continuous weights. From the viewpoint of hardware implementation, a simplified update rule is desirable. Although simplified STDP with stochastic binary synapses was proposed previously, we find that it leads to degradation of memory maintenance during learning, which is unfavourable for unsupervised online learning. In this work, we propose a stochastic binary synaptic model where the cumulative probability of the weight change evolves in a sigmoidal fashion with potentiation or depression trials, which can be implemented using a pair of switching devices consisting of serially connected multiple binary memristors. As a benchmark test we perform simulations of unsupervised learning of MNIST images with a two-layer network and show that simplified STDP in combination with this model can outperform conventional rules with continuous weights not only in memory maintenance but also in recognition accuracy. Our method achieves 97.3% in recognition accuracy, which is higher than that reported with standard STDP in the same framework. We also show that the high performance of our learning rule is robust against device-to-device variability of the memristor's probabilistic behaviour.
format	article
author	Yoshifumi Nishi Kumiko Nomura Takao Marukame Koichi Mizushima
author_facet	Yoshifumi Nishi Kumiko Nomura Takao Marukame Koichi Mizushima
author_sort	Yoshifumi Nishi
title	Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity
title_short	Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity
title_full	Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity
title_fullStr	Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity
title_full_unstemmed	Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity
title_sort	stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/1b42b9540ff943b4805ab44b662f914c
work_keys_str_mv	AT yoshifuminishi stochasticbinarysynapseshavingsigmoidalcumulativedistributionfunctionsforunsupervisedlearningwithspiketimingdependentplasticity AT kumikonomura stochasticbinarysynapseshavingsigmoidalcumulativedistributionfunctionsforunsupervisedlearningwithspiketimingdependentplasticity AT takaomarukame stochasticbinarysynapseshavingsigmoidalcumulativedistributionfunctionsforunsupervisedlearningwithspiketimingdependentplasticity AT koichimizushima stochasticbinarysynapseshavingsigmoidalcumulativedistributionfunctionsforunsupervisedlearningwithspiketimingdependentplasticity
_version_	1718378913515700224

Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity

Ejemplares similares