Spiking Neural Network Using Synaptic Transistors and Neuron Circuits for Pattern Recognition With Noisy Images

We demonstrate the hardware implementation of spiking neural network (SNN) with synaptic transistors and neuron circuits. The method of conversion from software fully-connected network (FCN) to hardware SNN with little degradation is discussed. The degradation of classification accuracy is analyzed...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE electron device letters Jg. 39; H. 4; S. 630 - 633
Hauptverfasser:	Kim, Hyungjin, Hwang, Sungmin, Park, Jungjin, Yun, Sangdoo, Lee, Jong-Ho, Park, Byung-Gook
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	IEEE 01.04.2018
Schlagworte:	Biological neural networks Degradation Denoising autoencoder (DAE) Hardware neuromorphic system neuron circuit Neurons Noise measurement pattern recognition Software spiking neural network (SNN) synaptic device Transistors
ISSN:	0741-3106, 1558-0563
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Abstract	We demonstrate the hardware implementation of spiking neural network (SNN) with synaptic transistors and neuron circuits. The method of conversion from software fully-connected network (FCN) to hardware SNN with little degradation is discussed. The degradation of classification accuracy is analyzed in terms of device variation and noisy images. In addition, the accuracy degradation is significantly improved by stacking denoising autoencoder (DAE) layer. FCN–SNN conversion with very little performance drop is demonstrated using weight normalization, and SNN with DAE layer shows a great tolerance to input image noise.
AbstractList	We demonstrate the hardware implementation of spiking neural network (SNN) with synaptic transistors and neuron circuits. The method of conversion from software fully-connected network (FCN) to hardware SNN with little degradation is discussed. The degradation of classification accuracy is analyzed in terms of device variation and noisy images. In addition, the accuracy degradation is significantly improved by stacking denoising autoencoder (DAE) layer. FCN–SNN conversion with very little performance drop is demonstrated using weight normalization, and SNN with DAE layer shows a great tolerance to input image noise.
Author	Lee, Jong-Ho Park, Jungjin Park, Byung-Gook Kim, Hyungjin Hwang, Sungmin Yun, Sangdoo
Author_xml	– sequence: 1 givenname: Hyungjin orcidid: 0000-0002-4834-6882 surname: Kim fullname: Kim, Hyungjin email: hyungjin@ece.ucsb.edu organization: Department of Electrical and Computer Engineering, University of California, Santa Barbara, CA, USA – sequence: 2 givenname: Sungmin surname: Hwang fullname: Hwang, Sungmin organization: Department of Electrical and Computer Engineering and the Inter-University Semiconductor Research Center, Seoul National University, Seoul, South Korea – sequence: 3 givenname: Jungjin surname: Park fullname: Park, Jungjin organization: Department of Electrical and Computer Engineering and the Inter-University Semiconductor Research Center, Seoul National University, Seoul, South Korea – sequence: 4 givenname: Sangdoo surname: Yun fullname: Yun, Sangdoo organization: Department of Electrical and Computer Engineering and the Inter-University Semiconductor Research Center, Seoul National University, Seoul, South Korea – sequence: 5 givenname: Jong-Ho orcidid: 0000-0003-3559-9802 surname: Lee fullname: Lee, Jong-Ho organization: Department of Electrical and Computer Engineering and the Inter-University Semiconductor Research Center, Seoul National University, Seoul, South Korea – sequence: 6 givenname: Byung-Gook orcidid: 0000-0002-2962-2458 surname: Park fullname: Park, Byung-Gook email: bgpark@snu.ac.kr organization: Department of Electrical and Computer Engineering and the Inter-University Semiconductor Research Center, Seoul National University, Seoul, South Korea
BookMark	eNp9kMFOAjEQhhuDiYjeTbz0BRan291u92gQlYSgEYjHzWzpYgVa0pYY3t5FiAcPnv5k8n8zme-SdKyzmpAbBn3GoLwbDx_6KTDZTyWUQrAz0mV5LhPIBe-QLhQZSzgDcUEuQ_gEYFlWZF3ipluzMnZJJ3rncd1G_HJ-RefhMJzuLW6jUXTm0QYTovOBol38tJ2lA-PVzsRAG-fpK8aovaVvWrmlNdG0hXcTP-jEmbCnow0udbgi5w2ug74-ZY_MH4ezwXMyfnkaDe7HiUoFj0kBIKXmKisbBqrUIq9RYlFLgU3KkUEjsMYiLQQ0TSHLFAXKOi11Vi8Y5sB7BI57lXcheN1UW2826PcVg-ogrGqFVQdh1UlYi4g_iDIRD29Ej2b9H3h7BI3W-veO5JDKkvFvGSF8Cg
CODEN	EDLEDZ
CitedBy_id	crossref_primary_10_3390_app11156703 crossref_primary_10_1016_j_neucom_2019_09_030 crossref_primary_10_1039_D4MH00053F crossref_primary_10_1016_j_neucom_2024_127653 crossref_primary_10_3389_fnano_2021_826232 crossref_primary_10_1016_j_chaos_2022_111999 crossref_primary_10_1039_D4MH01182A crossref_primary_10_1016_j_nanoen_2023_109169 crossref_primary_10_1109_ACCESS_2021_3108186 crossref_primary_10_3390_electronics10222784 crossref_primary_10_1063_1_5039544 crossref_primary_10_1007_s00530_025_01821_6 crossref_primary_10_1109_ACCESS_2020_3041734 crossref_primary_10_1002_aisy_202300456 crossref_primary_10_1109_LED_2023_3252579 crossref_primary_10_1016_j_cmpbup_2024_100171 crossref_primary_10_3390_biomimetics8050432 crossref_primary_10_1039_D5MH00086F crossref_primary_10_1109_ACCESS_2020_3046300 crossref_primary_10_1109_TED_2022_3186965 crossref_primary_10_1109_LED_2024_3445970 crossref_primary_10_1088_1361_6528_ac805d crossref_primary_10_1007_s00521_021_05699_z crossref_primary_10_1016_j_chaos_2022_112106 crossref_primary_10_1016_j_sse_2019_02_011 crossref_primary_10_1109_LED_2023_3265065 crossref_primary_10_3390_sym13010087 crossref_primary_10_1039_D1RA06492D crossref_primary_10_3390_ma13010166 crossref_primary_10_1007_s00530_024_01480_z crossref_primary_10_3390_math11092015 crossref_primary_10_1145_3304103 crossref_primary_10_1002_aisy_202100273 crossref_primary_10_1038_s41467_019_13827_6 crossref_primary_10_1038_s41467_022_28303_x crossref_primary_10_1063_5_0021626 crossref_primary_10_1109_TED_2020_3001492 crossref_primary_10_3390_electronics9010013 crossref_primary_10_3390_biomimetics9090578 crossref_primary_10_1016_j_chaos_2022_111813 crossref_primary_10_1109_LED_2022_3149029 crossref_primary_10_3390_math10010148 crossref_primary_10_1088_2053_1583_ab23ba crossref_primary_10_1002_adfm_202405544 crossref_primary_10_1016_j_chaos_2021_111587 crossref_primary_10_3390_s23239612 crossref_primary_10_3389_fnins_2020_00423 crossref_primary_10_1007_s10489_024_05299_z crossref_primary_10_1002_adma_202413916 crossref_primary_10_1016_j_jfranklin_2020_08_015 crossref_primary_10_1002_aisy_202200210 crossref_primary_10_3390_s21134462 crossref_primary_10_1016_j_neucom_2020_11_016 crossref_primary_10_1109_LED_2019_2958623 crossref_primary_10_1002_admt_202400063
Cites_doi	10.1038/nature14441 10.1088/1361-6528/aa86f8 10.1109/LED.2016.2521863 10.1145/2851613.2851724 10.1109/IJCNN.2016.7727303 10.1016/S0893-6080(97)00011-7 10.1109/LED.2017.2731859 10.1002/aelm.201600090 10.1109/LED.2016.2570279 10.1109/LED.2016.2517080 10.1109/TNNLS.2017.2778940 10.1109/TED.2004.834915 10.1142/S0129065712500141 10.3389/fncom.2015.00099 10.1109/5.58356 10.1021/acsnano.6b07894 10.1021/acsami.7b11191 10.1109/IEDM.2016.7838433 10.1109/TED.2017.2685519 10.1109/TED.2016.2630925 10.1109/IJCNN.2015.7280696 10.1016/j.neucom.2010.03.021 10.1088/0957-4484/27/36/365204 10.1109/LED.2017.2713460 10.1109/TCSII.2015.2456372 10.5573/JSTS.2016.16.5.657
ContentType	Journal Article
DBID	97E RIA RIE AAYXX CITATION
DOI	10.1109/LED.2018.2809661
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef
DatabaseTitle	CrossRef
DatabaseTitleList
Database_xml	– sequence: 1 dbid: RIE name: IEEE/IET Electronic Library url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	1558-0563
EndPage	633
ExternalDocumentID	10_1109_LED_2018_2809661 8302891
Genre	orig-research
GrantInformation_xml	– fundername: Brain Korea 21 Plus Project in 2017 – fundername: ICT grantid: 2016M3A7B4910348 – fundername: Nanomaterial Technology Development Programthrough the National Research Foundation of Korea through the Ministry of Science funderid: 10.13039/501100003725
GroupedDBID	-~X .DC 0R~ 29I 4.4 5GY 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACIWK ACNCT AENEX AETIX AFFNX AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD HZ~ IBMZZ ICLAB IFIPE IFJZH IPLJI JAVBF LAI M43 O9- OCL P2P RIA RIE RNS TAE TN5 TWZ VH1 AAYXX CITATION
ID	FETCH-LOGICAL-c263t-70088e3c49f10c9e65ba8a7b86af23a10f6aba72760ff7892a6a8b29e4bd1a503
IEDL.DBID	RIE
ISICitedReferencesCount	71
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000428689000043&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0741-3106
IngestDate	Tue Nov 18 19:37:56 EST 2025 Sat Nov 29 03:57:18 EST 2025 Wed Aug 27 02:51:58 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	4
Language	English
License	https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c263t-70088e3c49f10c9e65ba8a7b86af23a10f6aba72760ff7892a6a8b29e4bd1a503
ORCID	0000-0002-4834-6882 0000-0002-2962-2458 0000-0003-3559-9802
PageCount	4
ParticipantIDs	crossref_primary_10_1109_LED_2018_2809661 ieee_primary_8302891 crossref_citationtrail_10_1109_LED_2018_2809661
PublicationCentury	2000
PublicationDate	2018-April 2018-4-00
PublicationDateYYYYMMDD	2018-04-01
PublicationDate_xml	– month: 04 year: 2018 text: 2018-April
PublicationDecade	2010
PublicationTitle	IEEE electron device letters
PublicationTitleAbbrev	LED
PublicationYear	2018
Publisher	IEEE
Publisher_xml	– name: IEEE
References	ref13 ref12 ref15 ref14 ref11 ref10 ref1 ref17 ref16 ref18 vincent (ref20) 2010; 11 xie (ref21) 2012 prezioso (ref2) 2015; 521 ref24 ref23 ref26 ref25 ref22 ref28 ref27 ref29 ref8 ref7 ref9 ref4 ref3 ref6 ref5 o’connor (ref19) 2016
References_xml	– volume: 11 start-page: 3371 year: 2010 ident: ref20 article-title: Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion publication-title: J Mach Learn Res – volume: 521 start-page: 61 year: 2015 ident: ref2 article-title: Training and operation of an integrated neuromorphic network based on metal-oxide memristors publication-title: Nature doi: 10.1038/nature14441 – ident: ref24 doi: 10.1088/1361-6528/aa86f8 – ident: ref23 doi: 10.1109/LED.2016.2521863 – ident: ref12 doi: 10.1145/2851613.2851724 – start-page: 341 year: 2012 ident: ref21 article-title: Image denoising and inpainting with deep neural networks publication-title: Proc Adv Neural Inf Process Syst (NIPS) – ident: ref13 doi: 10.1109/IJCNN.2016.7727303 – ident: ref9 doi: 10.1016/S0893-6080(97)00011-7 – ident: ref18 doi: 10.1109/LED.2017.2731859 – ident: ref3 doi: 10.1002/aelm.201600090 – ident: ref4 doi: 10.1109/LED.2016.2570279 – ident: ref5 doi: 10.1109/LED.2016.2517080 – ident: ref17 doi: 10.1109/TNNLS.2017.2778940 – ident: ref27 doi: 10.1109/TED.2004.834915 – ident: ref11 doi: 10.1142/S0129065712500141 – ident: ref14 doi: 10.3389/fncom.2015.00099 – ident: ref1 doi: 10.1109/5.58356 – ident: ref16 doi: 10.1021/acsnano.6b07894 – ident: ref8 doi: 10.1021/acsami.7b11191 – ident: ref22 doi: 10.1109/IEDM.2016.7838433 – ident: ref26 doi: 10.1109/TED.2017.2685519 – ident: ref28 doi: 10.1109/TED.2016.2630925 – ident: ref29 doi: 10.1109/IJCNN.2015.7280696 – ident: ref10 doi: 10.1016/j.neucom.2010.03.021 – ident: ref6 doi: 10.1088/0957-4484/27/36/365204 – ident: ref7 doi: 10.1109/LED.2017.2713460 – year: 2016 ident: ref19 publication-title: Deep spiking networks – ident: ref15 doi: 10.1109/TCSII.2015.2456372 – ident: ref25 doi: 10.5573/JSTS.2016.16.5.657
SSID	ssj0014474
Score	2.5027983
Snippet	We demonstrate the hardware implementation of spiking neural network (SNN) with synaptic transistors and neuron circuits. The method of conversion from...
SourceID	crossref ieee
SourceType	Enrichment Source Index Database Publisher
StartPage	630
SubjectTerms	Biological neural networks Degradation Denoising autoencoder (DAE) Hardware neuromorphic system neuron circuit Neurons Noise measurement pattern recognition Software spiking neural network (SNN) synaptic device Transistors
Title	Spiking Neural Network Using Synaptic Transistors and Neuron Circuits for Pattern Recognition With Noisy Images
URI	https://ieeexplore.ieee.org/document/8302891
Volume	39
WOSCitedRecordID	wos000428689000043&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEEE/IET Electronic Library customDbUrl: eissn: 1558-0563 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014474 issn: 0741-3106 databaseCode: RIE dateStart: 19800101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1bS8MwFA5z-KAP3qY4b-TBF8G6rE1zeZS5oSBDvODeymmaYkHbsXaC_94krUVBBN9KOIHSr5xL8p3vIHRKdeIrpqgHXBKPJinz4jQknjCLIaMa_ADcsAk-nYrZTN510HnbC6O1duQzfWEf3V1-UqilPSobWK0qYVvVVzhnda9We2NAaa24bCKk8SukvZIkcnA7vrIcLnHhC5Ows-GPEPRtpooLKZPN_73MFtpoUkd8WWO9jTo630Hr3wQFe6h4mGf27BtbzQ1jO61J3tgRA_DDRw7GQyjsApTTBykx5ImzLnI8yhZqmVUlNoksvnPCmzm-_2IYGYPnrHrB0yIrP_DNm3FE5S56mowfR9deM1LBUz4LKo-bkC90oKhMh0RJzcIYBPBYMEgNKkOSMojB5DSMpCkX0gcGIvalpnEyhJAEe6ibF7neR1hKoKZcohCKhCY8AMUlhKbaNgmD8IOwjwZfXzlSjd64HXvxGrm6g8jI4BJZXKIGlz46a3fMa62NP2x7FpLWrkHj4PflQ7RmN9eMmyPUrRZLfYxW1XuVlYsT9yN9AoilxpM
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8QwEA6LCurBt_g2By-CdbNtmiZHWZVdXIv4QG9lmqZY0Fa2XWH_vUnaLQoieCthWkq_Mo_km28QOqEqcSWT1IFAEIcmKXPi1CcO14s-owpcD-ywiSAM-cuLuOugs7YXRillyWfq3Fzas_ykkBOzVdY1WlXctKrPm8lZTbdWe2ZAaa25rGOk9iykPZQkoju6ujQsLn7ucp2ys96PIPRtqooNKter_3udNbTSJI_4okZ7HXVUvoGWv0kKbqLi4SMzu9_YqG5o27CmeWNLDcAP0xy0j5DYhiirEFJiyBNrXeS4n43lJKtKrFNZfGelN3N8P-MYaYPnrHrFYZGVUzx8166o3EJP11eP_YHTDFVwpMu8ygl00OfKk1SkPSKFYn4MHIKYM0g1Lj2SMohBZzWMpGnAhQsMeOwKReOkBz7xttFcXuRqB2EhgOqCiYLPE5oEHshAgK_rbZ0ycNfzd1F39pUj2SiOm8EXb5GtPIiINC6RwSVqcNlFp-0dH7Xaxh-2mwaS1q5BY-_35WO0OHi8HUWjYXizj5bMg2r-zQGaq8YTdYgW5GeVleMj-1N9AQxCydw
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Spiking+Neural+Network+Using+Synaptic+Transistors+and+Neuron+Circuits+for+Pattern+Recognition+With+Noisy+Images&rft.jtitle=IEEE+electron+device+letters&rft.au=Kim%2C+Hyungjin&rft.au=Hwang%2C+Sungmin&rft.au=Park%2C+Jungjin&rft.au=Yun%2C+Sangdoo&rft.date=2018-04-01&rft.issn=0741-3106&rft.eissn=1558-0563&rft.volume=39&rft.issue=4&rft.spage=630&rft.epage=633&rft_id=info:doi/10.1109%2FLED.2018.2809661&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_LED_2018_2809661
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0741-3106&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0741-3106&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0741-3106&client=summon