Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits

Memristive devices used in neuromorphic computing typically need to be accessed using transistors, adding circuit complexity and size. In this work, the authors demonstrate a neural network using a transistor-free passive memristor crossbar array, offering potential circuit miniaturisation and energ...

Full description

Bibliographic Details
Main Authors: F. Merrikh Bayat, M. Prezioso, B. Chakrabarti, H. Nili, I. Kataeva, D. Strukov
Format: Article
Language:English
Published: Nature Publishing Group 2018-06-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-018-04482-4
id doaj-b43ad9a1d9064461ab060bcd4cf39d27
record_format Article
spelling doaj-b43ad9a1d9064461ab060bcd4cf39d272021-05-11T10:22:57ZengNature Publishing GroupNature Communications2041-17232018-06-01911710.1038/s41467-018-04482-4Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuitsF. Merrikh Bayat0M. Prezioso1B. Chakrabarti2H. Nili3I. Kataeva4D. Strukov5Electrical and Computer Engineering Department, University of CaliforniaElectrical and Computer Engineering Department, University of CaliforniaElectrical and Computer Engineering Department, University of CaliforniaElectrical and Computer Engineering Department, University of CaliforniaDENSO CORPElectrical and Computer Engineering Department, University of CaliforniaMemristive devices used in neuromorphic computing typically need to be accessed using transistors, adding circuit complexity and size. In this work, the authors demonstrate a neural network using a transistor-free passive memristor crossbar array, offering potential circuit miniaturisation and energy savings.https://doi.org/10.1038/s41467-018-04482-4
collection DOAJ
language English
format Article
sources DOAJ
author F. Merrikh Bayat
M. Prezioso
B. Chakrabarti
H. Nili
I. Kataeva
D. Strukov
spellingShingle F. Merrikh Bayat
M. Prezioso
B. Chakrabarti
H. Nili
I. Kataeva
D. Strukov
Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits
Nature Communications
author_facet F. Merrikh Bayat
M. Prezioso
B. Chakrabarti
H. Nili
I. Kataeva
D. Strukov
author_sort F. Merrikh Bayat
title Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits
title_short Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits
title_full Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits
title_fullStr Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits
title_full_unstemmed Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits
title_sort implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits
publisher Nature Publishing Group
series Nature Communications
issn 2041-1723
publishDate 2018-06-01
description Memristive devices used in neuromorphic computing typically need to be accessed using transistors, adding circuit complexity and size. In this work, the authors demonstrate a neural network using a transistor-free passive memristor crossbar array, offering potential circuit miniaturisation and energy savings.
url https://doi.org/10.1038/s41467-018-04482-4
work_keys_str_mv AT fmerrikhbayat implementationofmultilayerperceptronnetworkwithhighlyuniformpassivememristivecrossbarcircuits
AT mprezioso implementationofmultilayerperceptronnetworkwithhighlyuniformpassivememristivecrossbarcircuits
AT bchakrabarti implementationofmultilayerperceptronnetworkwithhighlyuniformpassivememristivecrossbarcircuits
AT hnili implementationofmultilayerperceptronnetworkwithhighlyuniformpassivememristivecrossbarcircuits
AT ikataeva implementationofmultilayerperceptronnetworkwithhighlyuniformpassivememristivecrossbarcircuits
AT dstrukov implementationofmultilayerperceptronnetworkwithhighlyuniformpassivememristivecrossbarcircuits
_version_ 1721448313303072768