Versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimization

Versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimization

Providing efficient and scalable specialized hardware for stochastic neural networks remains a challenge. Here, the authors propose a fast, energy-efficient and scalable stochastic dot-product circuit that may use either of two types of memory devices – metal-oxide memristors and floating-gate memor...

Full description

Bibliographic Details
Main Authors: M. R. Mahmoodi, M. Prezioso, D. B. Strukov
Format: Article
Language:English
Published: Nature Publishing Group 2019-11-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-019-13103-7
id doaj-75cd7755a38947a49505fa2462612ba9
record_format Article
spelling doaj-75cd7755a38947a49505fa2462612ba92021-05-11T12:23:57ZengNature Publishing GroupNature Communications2041-17232019-11-0110111010.1038/s41467-019-13103-7Versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimizationM. R. Mahmoodi0M. Prezioso1D. B. Strukov2Electrical and Computer Engineering Department, University of California Santa BarbaraElectrical and Computer Engineering Department, University of California Santa BarbaraElectrical and Computer Engineering Department, University of California Santa BarbaraProviding efficient and scalable specialized hardware for stochastic neural networks remains a challenge. Here, the authors propose a fast, energy-efficient and scalable stochastic dot-product circuit that may use either of two types of memory devices – metal-oxide memristors and floating-gate memories.https://doi.org/10.1038/s41467-019-13103-7
collection DOAJ
language English
format Article
sources DOAJ
author M. R. Mahmoodi
M. Prezioso
D. B. Strukov
spellingShingle M. R. Mahmoodi
M. Prezioso
D. B. Strukov
Versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimization
Nature Communications
author_facet M. R. Mahmoodi
M. Prezioso
D. B. Strukov
author_sort M. R. Mahmoodi
title Versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimization
title_short Versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimization
title_full Versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimization
title_fullStr Versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimization
title_full_unstemmed Versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimization
title_sort versatile stochastic dot product circuits based on nonvolatile memories for high performance neurocomputing and neurooptimization
publisher Nature Publishing Group
series Nature Communications
issn 2041-1723
publishDate 2019-11-01
description Providing efficient and scalable specialized hardware for stochastic neural networks remains a challenge. Here, the authors propose a fast, energy-efficient and scalable stochastic dot-product circuit that may use either of two types of memory devices – metal-oxide memristors and floating-gate memories.
url https://doi.org/10.1038/s41467-019-13103-7
work_keys_str_mv AT mrmahmoodi versatilestochasticdotproductcircuitsbasedonnonvolatilememoriesforhighperformanceneurocomputingandneurooptimization
AT mprezioso versatilestochasticdotproductcircuitsbasedonnonvolatilememoriesforhighperformanceneurocomputingandneurooptimization
AT dbstrukov versatilestochasticdotproductcircuitsbasedonnonvolatilememoriesforhighperformanceneurocomputingandneurooptimization
_version_ 1721444930708045824

Similar Items