TCL: an ANN-to-SNN Conversion with Trainable Clipping Layers

Spiking-neural-networks (SNNs) are promising at edge devices since the event-driven operations of SNNs provides significantly lower power compared to analog-neural-networks (ANNs). Although it is difficult to efficiently train SNNs, many techniques to convert trained ANNs to SNNs have been developed...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2021 58th ACM/IEEE Design Automation Conference (DAC) S. 793 - 798
Hauptverfasser: Ho, Nguyen-Dong, Chang, Ik-Joon
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 05.12.2021
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Spiking-neural-networks (SNNs) are promising at edge devices since the event-driven operations of SNNs provides significantly lower power compared to analog-neural-networks (ANNs). Although it is difficult to efficiently train SNNs, many techniques to convert trained ANNs to SNNs have been developed. However, after the conversion, a trade-off relation between accuracy and latency exists in SNNs, causing considerable latency in large size datasets such as ImageNet. We present a technique, named as TCL, to alleviate the trade-off problem, enabling the accuracy of 73.87% (VGG-16) and 70.37% (ResNet-34) for ImageNet with the moderate latency of 250 cycles in SNNs.
AbstractList Spiking-neural-networks (SNNs) are promising at edge devices since the event-driven operations of SNNs provides significantly lower power compared to analog-neural-networks (ANNs). Although it is difficult to efficiently train SNNs, many techniques to convert trained ANNs to SNNs have been developed. However, after the conversion, a trade-off relation between accuracy and latency exists in SNNs, causing considerable latency in large size datasets such as ImageNet. We present a technique, named as TCL, to alleviate the trade-off problem, enabling the accuracy of 73.87% (VGG-16) and 70.37% (ResNet-34) for ImageNet with the moderate latency of 250 cycles in SNNs.
Author Chang, Ik-Joon
Ho, Nguyen-Dong
Author_xml – sequence: 1
  givenname: Nguyen-Dong
  surname: Ho
  fullname: Ho, Nguyen-Dong
  email: donghn@khu.ac.kr
  organization: Kyung Hee University
– sequence: 2
  givenname: Ik-Joon
  surname: Chang
  fullname: Chang, Ik-Joon
  email: ichang@khu.ac.kr
  organization: Kyung Hee University
BookMark eNotj81KxDAcxCMoqGufQIS8QGvSfIuXElddKN2D9bwk7b8aqGlpi7Jvb8W9zMzhxzBzjc7jEAGhO0oySom5fyos1UTxLCc5zYzQMpfyDCVGaSql4CxXnFyiZJ6DJ5IIzVe9Qo-1LR-wi7ioqnQZ0reqwnaI3zDNYYj4JyyfuJ5ciM73gG0fxjHED1y640rcoIvO9TMkJ9-g9-dtbV_Tcv-ys0WZulyrJeVUwV9qjdGagddCuHUeePDSE-ZaYEYT0rK2aTppOspArUQrROMYlx3boNv_3gAAh3EKX246Hk4f2S-5CUg9
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/DAC18074.2021.9586266
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9781665432740
1665432748
EndPage 798
ExternalDocumentID 9586266
Genre orig-research
GrantInformation_xml – fundername: National Research Foundation of Korea
  funderid: 10.13039/501100003725
GroupedDBID 6IE
6IH
ACM
ALMA_UNASSIGNED_HOLDINGS
CBEJK
RIE
RIO
ID FETCH-LOGICAL-a287t-417ea287d99883eb855a862ebeb6b03ade39800d3dccf69f13e75a8d55ca346f3
IEDL.DBID RIE
ISICitedReferencesCount 37
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000766079700133&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:28:29 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a287t-417ea287d99883eb855a862ebeb6b03ade39800d3dccf69f13e75a8d55ca346f3
PageCount 6
ParticipantIDs ieee_primary_9586266
PublicationCentury 2000
PublicationDate 2021-Dec.-5
PublicationDateYYYYMMDD 2021-12-05
PublicationDate_xml – month: 12
  year: 2021
  text: 2021-Dec.-5
  day: 05
PublicationDecade 2020
PublicationTitle 2021 58th ACM/IEEE Design Automation Conference (DAC)
PublicationTitleAbbrev DAC
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssib060584060
Score 2.4465046
Snippet Spiking-neural-networks (SNNs) are promising at edge devices since the event-driven operations of SNNs provides significantly lower power compared to...
SourceID ieee
SourceType Publisher
StartPage 793
SubjectTerms ANN-to-SNN conversion
Clocks
Design automation
Limiting
Neural networks
Spiking Neural Network
trainable clipping layer
Training
Title TCL: an ANN-to-SNN Conversion with Trainable Clipping Layers
URI https://ieeexplore.ieee.org/document/9586266
WOSCitedRecordID wos000766079700133&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
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFH5sw4MnlU38TQ4ezdY2aZOIl1EdHkYZOGG3kaYvIkgns_PvN8nqRPDiLYSWNC-F73sv73sP4DpJBDoc0DQThlNuTUS1wog6V8GW1kpWZmVoNiGKQi4WataBm50WBhFD8hkO_TDc5Vcrs_GhspFKPf_OutAVQmy1Wt__jr_dc9gUtSKdOFKj-3Ee-1IvzglM4mH77q8mKgFDJgf_W_0QBj9iPDLbwcwRdLDuw908n94SXZNxUdBmRZ-KguQ-gTxEv4iPrpL5ulVGkfwtlGF4IVPtGfYAnicP8_yRto0QqHYOTUN5LNCPKucbSYalTFPtvsXZ31kyYrpCphzxq1hljM2UjRkK90SVpkYznll2DL16VeMJEBGpSnIH26ZMOBqtE88ImeMxpeSSm1Po-50v37e1Lpbtps_-nj6HfW_ckN6RXkCvWW_wEvbMZ_P6sb4KB_QFHG-P9A
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFH7MKehJZRN_m4NHs7VN0jbiZVTHxBoGVthtpGkqgnRSO_9-k6xOBC_eQmlo8xL4vvfyvvcALoMg0gYHJA4jRTEtlYcl1x42rkKZl2VM8jB3zSYiIeLZjE87cLXWwmitXfKZHtihu8svFmppQ2VDziz_Djdgk1Ea-Cu11vfpsfd7Bp28Vqbje3x4O0p8W-zFuIGBP2hn_2qj4lBkvPu_7-9B_0eOh6ZroNmHjq56cJMl6TWSFRoJgZsFfhICJTaF3MW_kI2voqxutVEoeXOFGF5QKi3H7sPz-C5LJrhthYClcWkaTP1I21FhvKOY6DxmTJp_MTtgbOkRWWjCDfUrSKFUGfLSJzoybxSMKUloWJID6FaLSh8CijxexNQAt8oDqpWUgeWExDCZPKYxVUfQsyufv6-qXczbRR___fgCtifZYzpP78XDCexYQ7tkD3YK3aZe6jPYUp_N60d97jbrC86Fkzs
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%3Abook&rft.genre=proceeding&rft.title=2021+58th+ACM%2FIEEE+Design+Automation+Conference+%28DAC%29&rft.atitle=TCL%3A+an+ANN-to-SNN+Conversion+with+Trainable+Clipping+Layers&rft.au=Ho%2C+Nguyen-Dong&rft.au=Chang%2C+Ik-Joon&rft.date=2021-12-05&rft.pub=IEEE&rft.spage=793&rft.epage=798&rft_id=info:doi/10.1109%2FDAC18074.2021.9586266&rft.externalDocID=9586266