File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Analog error correcting codes for defect tolerant matrix multiplication in crossbars

TitleAnalog error correcting codes for defect tolerant matrix multiplication in crossbars
Authors
Issue Date2020
PublisherIEEE. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000245
Citation
2020 IEEE International Electron Devices Meeting (IEDM), San Francisco, CA, USA, 12-18 December 2020, p. 36.6.1-36.6.4 How to Cite?
AbstractDespite great promises shown in the laboratory environment, memristor crossbar, or non-volatile resistive analog memory, based matrix multiplication accelerators suffer from unexpected computing errors, limiting opportunities to replace main-stream digital systems. While many previously demonstrated applications, such as neural networks, are tolerant of small errors, they are challenged by any significant outliers, which must be detected and corrected. Herein, we experimentally demonstrate an analog Error Correcting Code (ECC) scheme that considerably reduces the chance of substantial errors, by detecting and correcting errors with minimum hardware overhead. Different from well-known digital ECC in communication and memory, this analog version can tolerate small errors while detecting and correcting those over a predefined threshold. With this scheme, we can recover the MNIST handwritten digit classification accuracy experimentally from 90.31% to 96.21% in the event an array builds up shorted devices and from 73.12% to 97.36% when current noise is injected. For applications where high reliability and compute precision are demanded, such as in high-performance and scientific computing, we expect the schemes shown here to make analog computing more feasible.
Persistent Identifierhttp://hdl.handle.net/10722/305968
ISSN
2023 SCImago Journal Rankings: 1.047
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorLi, C-
dc.contributor.authorRoth, RM-
dc.contributor.authorGraves, C-
dc.contributor.authorSheng, X-
dc.contributor.authorStrachan, JP-
dc.date.accessioned2021-10-20T10:16:56Z-
dc.date.available2021-10-20T10:16:56Z-
dc.date.issued2020-
dc.identifier.citation2020 IEEE International Electron Devices Meeting (IEDM), San Francisco, CA, USA, 12-18 December 2020, p. 36.6.1-36.6.4-
dc.identifier.issn0163-1918-
dc.identifier.urihttp://hdl.handle.net/10722/305968-
dc.description.abstractDespite great promises shown in the laboratory environment, memristor crossbar, or non-volatile resistive analog memory, based matrix multiplication accelerators suffer from unexpected computing errors, limiting opportunities to replace main-stream digital systems. While many previously demonstrated applications, such as neural networks, are tolerant of small errors, they are challenged by any significant outliers, which must be detected and corrected. Herein, we experimentally demonstrate an analog Error Correcting Code (ECC) scheme that considerably reduces the chance of substantial errors, by detecting and correcting errors with minimum hardware overhead. Different from well-known digital ECC in communication and memory, this analog version can tolerate small errors while detecting and correcting those over a predefined threshold. With this scheme, we can recover the MNIST handwritten digit classification accuracy experimentally from 90.31% to 96.21% in the event an array builds up shorted devices and from 73.12% to 97.36% when current noise is injected. For applications where high reliability and compute precision are demanded, such as in high-performance and scientific computing, we expect the schemes shown here to make analog computing more feasible.-
dc.languageeng-
dc.publisherIEEE. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000245-
dc.relation.ispartofIEEE International Electron Devices Meeting (IEDM)-
dc.rightsIEEE International Electron Devices Meeting (IEDM). Copyright © IEEE.-
dc.rights©2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.-
dc.titleAnalog error correcting codes for defect tolerant matrix multiplication in crossbars-
dc.typeConference_Paper-
dc.identifier.emailLi, C: canl@hku.hk-
dc.identifier.authorityLi, C=rp02706-
dc.description.naturepostprint-
dc.identifier.doi10.1109/IEDM13553.2020.9371978-
dc.identifier.scopuseid_2-s2.0-85102940012-
dc.identifier.hkuros327482-
dc.identifier.spage36.6.1-
dc.identifier.epage36.6.4-
dc.identifier.isiWOS:000717011600089-
dc.publisher.placeUnited States-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats