High-Accuracy Deep Belief Network: Fuzzy Neural Networks Using MRAM-Based Stochastic Neurons

Two-terminal magnetoresistive random access memory (MRAM) devices provide a recent approach to intrinsically realizing stochastic neuronal behavior in cognitive architectures such as restricted Boltzmann machines (RBMs) for deep belief networks (DBNs). MRAM-based DBNs have achieved substantial energ...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Hossein Pourmeidani, Ronald F. Demara
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Deep belief network (DBN) fuzzy system low energy barrier magnetic tunnel junction (MTJ) probabilistic spin logic device (p-bit) Computer engineering. Computer hardware TK7885-7895
Acceso en línea:	https://doaj.org/article/da222e16512b49659d0eb142e2eb88d5
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:da222e16512b49659d0eb142e2eb88d5
record_format	dspace
spelling	oai:doaj.org-article:da222e16512b49659d0eb142e2eb88d52021-11-18T00:11:26ZHigh-Accuracy Deep Belief Network: Fuzzy Neural Networks Using MRAM-Based Stochastic Neurons2329-923110.1109/JXCDC.2021.3117489https://doaj.org/article/da222e16512b49659d0eb142e2eb88d52021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9558846/https://doaj.org/toc/2329-9231Two-terminal magnetoresistive random access memory (MRAM) devices provide a recent approach to intrinsically realizing stochastic neuronal behavior in cognitive architectures such as restricted Boltzmann machines (RBMs) for deep belief networks (DBNs). MRAM-based DBNs have achieved substantial energy and area improvements in comparison with the prior DBN hardware implementations. However, MRAM-based DBNs suffer from low accuracy in image classification. In order to resolve this problem, we present a new DBN-fuzzy system based on the combination of MRAM-based DBNs and fuzzy systems in the interest of improving the accuracy of MRAM-based neural networks. First, the MRAM-based DBN is employed to identify the top recognition results with the highest probability. Second, the fuzzy system is utilized to obtain the top-1 recognition results. We assess the accuracy of neural networks on the MNIST dataset, finding that the top-1 accuracy of the DBN-fuzzy neural network is improved from 64% to 82% in comparison to the MRAM-based DBNs for a <inline-formula> <tex-math notation="LaTeX">$784\times10$ </tex-math></inline-formula> network.Hossein PourmeidaniRonald F. DemaraIEEEarticleDeep belief network (DBN)fuzzy systemlow energy barrier magnetic tunnel junction (MTJ)probabilistic spin logic device (p-bit)Computer engineering. Computer hardwareTK7885-7895ENIEEE Journal on Exploratory Solid-State Computational Devices and Circuits, Vol 7, Iss 2, Pp 125-131 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Deep belief network (DBN) fuzzy system low energy barrier magnetic tunnel junction (MTJ) probabilistic spin logic device (p-bit) Computer engineering. Computer hardware TK7885-7895
spellingShingle	Deep belief network (DBN) fuzzy system low energy barrier magnetic tunnel junction (MTJ) probabilistic spin logic device (p-bit) Computer engineering. Computer hardware TK7885-7895 Hossein Pourmeidani Ronald F. Demara High-Accuracy Deep Belief Network: Fuzzy Neural Networks Using MRAM-Based Stochastic Neurons
description	Two-terminal magnetoresistive random access memory (MRAM) devices provide a recent approach to intrinsically realizing stochastic neuronal behavior in cognitive architectures such as restricted Boltzmann machines (RBMs) for deep belief networks (DBNs). MRAM-based DBNs have achieved substantial energy and area improvements in comparison with the prior DBN hardware implementations. However, MRAM-based DBNs suffer from low accuracy in image classification. In order to resolve this problem, we present a new DBN-fuzzy system based on the combination of MRAM-based DBNs and fuzzy systems in the interest of improving the accuracy of MRAM-based neural networks. First, the MRAM-based DBN is employed to identify the top recognition results with the highest probability. Second, the fuzzy system is utilized to obtain the top-1 recognition results. We assess the accuracy of neural networks on the MNIST dataset, finding that the top-1 accuracy of the DBN-fuzzy neural network is improved from 64% to 82% in comparison to the MRAM-based DBNs for a <inline-formula> <tex-math notation="LaTeX">$784\times10$ </tex-math></inline-formula> network.
format	article
author	Hossein Pourmeidani Ronald F. Demara
author_facet	Hossein Pourmeidani Ronald F. Demara
author_sort	Hossein Pourmeidani
title	High-Accuracy Deep Belief Network: Fuzzy Neural Networks Using MRAM-Based Stochastic Neurons
title_short	High-Accuracy Deep Belief Network: Fuzzy Neural Networks Using MRAM-Based Stochastic Neurons
title_full	High-Accuracy Deep Belief Network: Fuzzy Neural Networks Using MRAM-Based Stochastic Neurons
title_fullStr	High-Accuracy Deep Belief Network: Fuzzy Neural Networks Using MRAM-Based Stochastic Neurons
title_full_unstemmed	High-Accuracy Deep Belief Network: Fuzzy Neural Networks Using MRAM-Based Stochastic Neurons
title_sort	high-accuracy deep belief network: fuzzy neural networks using mram-based stochastic neurons
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/da222e16512b49659d0eb142e2eb88d5
work_keys_str_mv	AT hosseinpourmeidani highaccuracydeepbeliefnetworkfuzzyneuralnetworksusingmrambasedstochasticneurons AT ronaldfdemara highaccuracydeepbeliefnetworkfuzzyneuralnetworksusingmrambasedstochasticneurons
_version_	1718425194236739584

High-Accuracy Deep Belief Network: Fuzzy Neural Networks Using MRAM-Based Stochastic Neurons

Ejemplares similares