Latency-driven design for FPGA-based convolutional neural networks

In recent years, Convolutional Neural Networks (ConvNets) have become the quintessential component of several state-of-the-art Artificial Intelligence tasks. Across the spectrum of applications, the performance needs vary significantly, from high-throughput image recognition to the very low-latency...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International Conference on Field-programmable Logic and Applications s. 1 - 8
Hlavní autoři: Venieris, Stylianos I., Bouganis, Christos-Savvas
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: Ghent University 01.09.2017
Témata:
ISSN:1946-1488
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract In recent years, Convolutional Neural Networks (ConvNets) have become the quintessential component of several state-of-the-art Artificial Intelligence tasks. Across the spectrum of applications, the performance needs vary significantly, from high-throughput image recognition to the very low-latency requirements of autonomous cars. In this context, FPGAs can provide a potential platform that can be optimally configured based on different performance requirements. However, with the increasing complexity of ConvNet models, the architectural design space becomes overwhelmingly large, asking for principled design flows that address the application-level needs. This paper presents a latency-driven design methodology for mapping ConvNets on FPGAs. The proposed design flow employs novel transformations over a Synchronous Dataflow-based modelling framework together with a latency-centric optimisation procedure in order to efficiently explore the design space targeting low-latency designs. Quantitative evaluation shows large improvements in latency when latency-driven optimisation is in place yielding designs that improve the latency of AlexNet by 73.54× and VGG16 by 5.61× over throughput-optimised designs.
AbstractList In recent years, Convolutional Neural Networks (ConvNets) have become the quintessential component of several state-of-the-art Artificial Intelligence tasks. Across the spectrum of applications, the performance needs vary significantly, from high-throughput image recognition to the very low-latency requirements of autonomous cars. In this context, FPGAs can provide a potential platform that can be optimally configured based on different performance requirements. However, with the increasing complexity of ConvNet models, the architectural design space becomes overwhelmingly large, asking for principled design flows that address the application-level needs. This paper presents a latency-driven design methodology for mapping ConvNets on FPGAs. The proposed design flow employs novel transformations over a Synchronous Dataflow-based modelling framework together with a latency-centric optimisation procedure in order to efficiently explore the design space targeting low-latency designs. Quantitative evaluation shows large improvements in latency when latency-driven optimisation is in place yielding designs that improve the latency of AlexNet by 73.54× and VGG16 by 5.61× over throughput-optimised designs.
Author Venieris, Stylianos I.
Bouganis, Christos-Savvas
Author_xml – sequence: 1
  givenname: Stylianos I.
  surname: Venieris
  fullname: Venieris, Stylianos I.
  email: stylianos.venieris10@imperial.ac.uk
– sequence: 2
  givenname: Christos-Savvas
  surname: Bouganis
  fullname: Bouganis, Christos-Savvas
  email: christos-savvas.bouganis@imperial.ac.uk
BookMark eNotj81KAzEYAKMo2NY-gHjZF8ia_02OtbhVWLAHPZds8q1E10SSbaVvr2jnMreBmaOLmCIgdENJzbih5q7ddjUjtKk1kUozfYbmhhjCiWCanaMZNUJhKrS-QstS3skvUjRaqhm67-wE0R2xz-EAsfJQwlushpSrdrtZ4d4W8JVL8ZDG_RRStGMVYZ__NH2n_FGu0eVgxwLLkxfotX14WT_i7nnztF51ONBGTlgpqQeroOm94c4A00CE89xIJhvuhdXe-d4TJj3nhgnee-ZcD855xyzVfIFu_7sBAHZfOXzafNydlvkPleRNBQ
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.23919/FPL.2017.8056828
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 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 9090304282
9789090304281
EISSN 1946-1488
EndPage 8
ExternalDocumentID 8056828
Genre orig-research
GroupedDBID 6IE
6IF
6IL
6IN
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
OCL
RIE
RIL
ID FETCH-LOGICAL-i175t-6658fa6e7bd93c9e28e04cd3952573d4a8dcdbd025d339243bd2ccbeccdc2a183
IEDL.DBID RIE
ISICitedReferencesCount 19
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000426989400073&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 02:28:39 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i175t-6658fa6e7bd93c9e28e04cd3952573d4a8dcdbd025d339243bd2ccbeccdc2a183
PageCount 8
ParticipantIDs ieee_primary_8056828
PublicationCentury 2000
PublicationDate 2017-Sept.
PublicationDateYYYYMMDD 2017-09-01
PublicationDate_xml – month: 09
  year: 2017
  text: 2017-Sept.
PublicationDecade 2010
PublicationTitle International Conference on Field-programmable Logic and Applications
PublicationTitleAbbrev FPL
PublicationYear 2017
Publisher Ghent University
Publisher_xml – name: Ghent University
SSID ssj0000547856
Score 2.0267925
Snippet In recent years, Convolutional Neural Networks (ConvNets) have become the quintessential component of several state-of-the-art Artificial Intelligence tasks....
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Biological system modeling
Computational modeling
Convolution
Feature extraction
Field programmable gate arrays
Machine learning
Space exploration
Title Latency-driven design for FPGA-based convolutional neural networks
URI https://ieeexplore.ieee.org/document/8056828
WOSCitedRecordID wos000426989400073&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/eLvHCXMwlV1LSwMxEB7a4sGTSiu-ycGj6WOzu0mOKq4eStmDQm9lNpNCQVrpQ_Dfm8kuFcGLp4TAsuSx-32Z-WYG4JYyTHVWoTSUaRkQwsuKPYXzylCO6IcGMRab0JOJmU5t2YK7fSyM9z6Kz3yfu9GXTyu3Y1PZwAS0DjeENrS11nWs1t6eMuTEVFleOy4TZUd2UJRj1m7pfvPcrwIqET-Ko_-9-Rh6P4F4otxDzAm0_LILD2Nkovslac2_KkFRhCEC-xRF-XwvGZhIsJy8OVb4LjhtZWyi6HvTg7fi6fXxRTalEOQi4PtW5oEozDH3uiKrnPWJ8cPUkbKczFRRioYcVRQIDKnAeFJVUeIc7w-5BMNnewqd5Wrpz0AYleuELLFJLa1GGpXLjHXWKEQMN7Jz6PL8Zx91totZM_WLv4cv4ZCXuFZdXUFnu975azhwn9vFZn0Tt-gb25-T7w
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEB5qFfSk0opv9-DR9LHZR3JUca24lj1U6K3MZlIQpJU-BP-9mexSEbx4SggsSx6735eZb2YArinGKI1LFIriVDiEsKJkT-G0VJQg2p5C9MUm0uFQjce6aMDNJhbGWuvFZ7bDXe_Lp7lZs6msqxxauxvCFmzHURT2q2itjUWlx6mp4qRyXYZS93U3K3JWb6Wd-slfJVQ8gmT7_3v3AbR_QvGCYgMyh9Cwsxbc5chU90vQgn9WAXkZRuD4Z5AVj7eCoYkCFpTXBwvfA05c6Rsv-1624TV7GN0PRF0MQbw5hF-JxFGFKSY2LUlLo22obC8yJDWnM5UUoSJDJTkKQ9JxnkiWFBrDO0QmRPfhHkFzNp_ZYwiUTNKQNLFRLSr7KUoTK220kojo7mQn0OL5Tz6qfBeTeuqnfw9fwe5g9JJP8qfh8xns8XJXGqxzaK4Wa3sBO-Zz9bZcXPrt-ga-mJc2
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=proceeding&rft.title=International+Conference+on+Field-programmable+Logic+and+Applications&rft.atitle=Latency-driven+design+for+FPGA-based+convolutional+neural+networks&rft.au=Venieris%2C+Stylianos+I.&rft.au=Bouganis%2C+Christos-Savvas&rft.date=2017-09-01&rft.pub=Ghent+University&rft.eissn=1946-1488&rft.spage=1&rft.epage=8&rft_id=info:doi/10.23919%2FFPL.2017.8056828&rft.externalDocID=8056828