File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: A memristive deep belief neural network based on silicon synapses

TitleA memristive deep belief neural network based on silicon synapses
Authors
Issue Date19-Dec-2022
PublisherNature Research
Citation
Nature Electronics, 2022, v. 5, n. 12, p. 870-880 How to Cite?
Abstract

Memristor-based neuromorphic computing could overcome the limitations of traditional von Neumann computing architectures-in which data are shuffled between separate memory and processing units-and improve the performance of deep neural networks. However, this will require accurate synaptic-like device performance, and memristors typically suffer from poor yield and a limited number of reliable conductance states. Here we report floating-gate memristive synaptic devices that are fabricated in a commercial complementary metal-oxide-semiconductor process. These silicon synapses offer analogue tunability, high endurance, long retention time, predictable cycling degradation, moderate device-to-device variation and high yield. They also provide two orders of magnitude higher energy efficiency for multiply-accumulate operations than graphics processing units. We use two 12 x 8 arrays of memristive devices for the in situ training of a 19 x 8 memristive restricted Boltzmann machine for pattern recognition via a gradient descent algorithm based on contrastive divergence. We then create a memristive deep belief neural network consisting of three memristive restricted Boltzmann machines. We test this system using the modified National Institute of Standards and Technology dataset, demonstrating a recognition accuracy of up to 97.05%.


Persistent Identifierhttp://hdl.handle.net/10722/337552
ISSN
2023 Impact Factor: 33.7
2023 SCImago Journal Rankings: 11.667
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorWang, Wei-
dc.contributor.authorDanial, Loai-
dc.contributor.authorLi, Yang-
dc.contributor.authorHerbelin, Eric-
dc.contributor.authorPikhay, Evgeny-
dc.contributor.authorRoizin, Yakov-
dc.contributor.authorHoffer, Barak-
dc.contributor.authorWang, Zhongrui-
dc.contributor.authorKvatinsky, Shahar -
dc.date.accessioned2024-03-11T10:21:47Z-
dc.date.available2024-03-11T10:21:47Z-
dc.date.issued2022-12-19-
dc.identifier.citationNature Electronics, 2022, v. 5, n. 12, p. 870-880-
dc.identifier.issn2520-1131-
dc.identifier.urihttp://hdl.handle.net/10722/337552-
dc.description.abstract<p>Memristor-based neuromorphic computing could overcome the limitations of traditional von Neumann computing architectures-in which data are shuffled between separate memory and processing units-and improve the performance of deep neural networks. However, this will require accurate synaptic-like device performance, and memristors typically suffer from poor yield and a limited number of reliable conductance states. Here we report floating-gate memristive synaptic devices that are fabricated in a commercial complementary metal-oxide-semiconductor process. These silicon synapses offer analogue tunability, high endurance, long retention time, predictable cycling degradation, moderate device-to-device variation and high yield. They also provide two orders of magnitude higher energy efficiency for multiply-accumulate operations than graphics processing units. We use two 12 x 8 arrays of memristive devices for the in situ training of a 19 x 8 memristive restricted Boltzmann machine for pattern recognition via a gradient descent algorithm based on contrastive divergence. We then create a memristive deep belief neural network consisting of three memristive restricted Boltzmann machines. We test this system using the modified National Institute of Standards and Technology dataset, demonstrating a recognition accuracy of up to 97.05%.<br></p>-
dc.languageeng-
dc.publisherNature Research-
dc.relation.ispartofNature Electronics-
dc.titleA memristive deep belief neural network based on silicon synapses-
dc.typeArticle-
dc.identifier.doi10.1038/s41928-022-00878-9-
dc.identifier.scopuseid_2-s2.0-85144244560-
dc.identifier.volume5-
dc.identifier.issue12-
dc.identifier.spage870-
dc.identifier.epage880-
dc.identifier.eissn2520-1131-
dc.identifier.isiWOS:000900800000004-
dc.identifier.issnl2520-1131-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats