Residual-INR: Communication Efficient on-Device Learning Using Implicit Neural Representation

Edge computing is a distributed computing paradigm that collects and processes data at or near the source of data generation. The on-device learning at edge relies on device-to-device wireless communication to facilitate real-time data sharing and collaborative decision-making among multiple devices...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Digest of technical papers - IEEE/ACM International Conference on Computer-Aided Design s. 1 - 9
Hlavní autoři: Chen, Hanqiu, Yao, Xuebin, Subedi, Pradeep, Hao, Cong
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: ACM 27.10.2024
Témata:
ISSN:1558-2434
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Edge computing is a distributed computing paradigm that collects and processes data at or near the source of data generation. The on-device learning at edge relies on device-to-device wireless communication to facilitate real-time data sharing and collaborative decision-making among multiple devices. This significantly improves the adaptability of the edge computing system to the changing environments. However, as the scale of the edge computing system is getting larger, communication among devices is becoming the bottleneck because of the limited bandwidth of wireless communication leads to large data transfer latency. To reduce the amount of device-to-device data transmission and accelerate on-device learning, in this paper, we propose Residual-INR, a fog computing-based communication-efficient on-device learning framework by utilizing implicit neural representation (INR) to compress images/videos into neural network weights. Residual-INR enhances data transfer efficiency by collecting JPEG images from edge devices, compressing them into INR format at the fog node, and redistributing them for on-device learning. By using a smaller INR for full image encoding and a separate object INR for high-quality object region reconstruction through residual encoding, our technique can reduce the encoding redundancy while maintaining the object quality. Residual-INR is a promising solution for edge on-device learning because it reduces data transmission by up to 5.16 \times across a network of 10 edge devices. It also facilitates CPU-free accelerated on-device learning, achieving up to 2.9 \times speedup without sacrificing accuracy. Our code is available at: https://github.com/sharc-lab/Residual-INR.
AbstractList Edge computing is a distributed computing paradigm that collects and processes data at or near the source of data generation. The on-device learning at edge relies on device-to-device wireless communication to facilitate real-time data sharing and collaborative decision-making among multiple devices. This significantly improves the adaptability of the edge computing system to the changing environments. However, as the scale of the edge computing system is getting larger, communication among devices is becoming the bottleneck because of the limited bandwidth of wireless communication leads to large data transfer latency. To reduce the amount of device-to-device data transmission and accelerate on-device learning, in this paper, we propose Residual-INR, a fog computing-based communication-efficient on-device learning framework by utilizing implicit neural representation (INR) to compress images/videos into neural network weights. Residual-INR enhances data transfer efficiency by collecting JPEG images from edge devices, compressing them into INR format at the fog node, and redistributing them for on-device learning. By using a smaller INR for full image encoding and a separate object INR for high-quality object region reconstruction through residual encoding, our technique can reduce the encoding redundancy while maintaining the object quality. Residual-INR is a promising solution for edge on-device learning because it reduces data transmission by up to 5.16 \times across a network of 10 edge devices. It also facilitates CPU-free accelerated on-device learning, achieving up to 2.9 \times speedup without sacrificing accuracy. Our code is available at: https://github.com/sharc-lab/Residual-INR.
Author Chen, Hanqiu
Hao, Cong
Subedi, Pradeep
Yao, Xuebin
Author_xml – sequence: 1
  givenname: Hanqiu
  surname: Chen
  fullname: Chen, Hanqiu
  email: hanqiu.chen@gatech.edu
  organization: Georgia Institute of Technology
– sequence: 2
  givenname: Xuebin
  surname: Yao
  fullname: Yao, Xuebin
  email: xuebin.yao@samsung.com
  organization: Samsung Semiconductor, Inc
– sequence: 3
  givenname: Pradeep
  surname: Subedi
  fullname: Subedi, Pradeep
  email: prad.subedi@samsung.com
  organization: Samsung Semiconductor, Inc
– sequence: 4
  givenname: Cong
  surname: Hao
  fullname: Hao, Cong
  email: callie.hao@gatech.edu
  organization: Georgia Institute of Technology
BookMark eNotjM1Kw0AYRUdRsNau3biYF0idb_7HndSqhVIh2KWU6eSLDCSTkB_BtzdVN_dszj3X5CI1CQm5BbYEkOpeaKOV0MsTtRFnZOGMs5IxA8wYcU5moJTNuBTyiiz6Ph6ZZmoSpJ2Rjxz7WIy-yja7_IGumroeUwx-iE2i67KMIWIaaJOyJ_yKAekWfZdi-qT7_rSbuq0mZ6A7HDtf0RzbDvvp8lu4IZelr3pc_HNO9s_r99Vrtn172awet5nnDoYsmIJJDYhSKQhgCpBQeHlEHpgtTdCOg3fSOK4F2kk-2tIJ6wSoSS2YmJO7v25ExEPbxdp33wcA4JoLIX4AcelVTA
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1145/3676536.3676673
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE/IET Electronic Library (IEL) (UW System Shared)
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
Discipline Engineering
EISBN 9798400710773
EISSN 1558-2434
EndPage 9
ExternalDocumentID 11126233
Genre orig-research
GrantInformation_xml – fundername: National Science Foundation
  grantid: ENG-ECCS2202310
  funderid: 10.13039/100000001
– fundername: Samsung
  grantid: ENG-ECCS2202310
  funderid: 10.13039/100004358
– fundername: Cisco
  grantid: ENG-ECCS2202310
  funderid: 10.13039/100004351
GroupedDBID 6IE
6IF
6IH
6IL
6IN
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
FEDTE
IEGSK
IJVOP
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-a291t-c7d0461ee4551c17d141da4be2c08f7c6921a9479263e87d0b8f9389315d14d03
IEDL.DBID RIE
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001479882200112&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Sep 03 07:09:38 EDT 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a291t-c7d0461ee4551c17d141da4be2c08f7c6921a9479263e87d0b8f9389315d14d03
OpenAccessLink https://dl.acm.org/doi/pdf/10.1145/3676536.3676673
PageCount 9
ParticipantIDs ieee_primary_11126233
PublicationCentury 2000
PublicationDate 2024-Oct.-27
PublicationDateYYYYMMDD 2024-10-27
PublicationDate_xml – month: 10
  year: 2024
  text: 2024-Oct.-27
  day: 27
PublicationDecade 2020
PublicationTitle Digest of technical papers - IEEE/ACM International Conference on Computer-Aided Design
PublicationTitleAbbrev ICCAD
PublicationYear 2024
Publisher ACM
Publisher_xml – name: ACM
SSID ssib060584048
ssj0020286
Score 1.9047651
Snippet Edge computing is a distributed computing paradigm that collects and processes data at or near the source of data generation. The on-device learning at edge...
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Data transfer
Edge computing
Efficient Communication
Fog Computing
Image coding
Image edge detection
Implicit Neural Representation(INR)
On-device Learning
Pipelines
Training
Transform coding
Video compression
Videos
Wireless communication
Title Residual-INR: Communication Efficient on-Device Learning Using Implicit Neural Representation
URI https://ieeexplore.ieee.org/document/11126233
WOSCitedRecordID wos001479882200112&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/eLvHCXMwlV1LSwMxEA62eNCLr4pvcvC67eaxeXjVFnsppSj0IiXJZqUgW2m3_n4z2W2tBw-eEkIIIVl2JjPzfR9C92lmiIWojWMSolXSJFaBiAAzPC1U4RzNo9iEHI3UdKrHDVg9YmG897H4zHehG3P5-cKtIVTWI4B3oYy1UEtKUYO1Nh8PpPd4yrcphPCoV6Lh8iE86wEzWcZEF1oBKuk7YirRlgyO_rmLY9T5QeXh8dbenKA9X56iwx1CwTP0NvGriK5KhqPJA_6F_sD9yBYR1saLMnny8IvADb3qO46lA3gY68vnFQbODvOBJ7FOtoEnlR30Oui_PD4njYBCYqgmVeJkDnzq3vPgFzkic8JJbrj11KWqkE5oSozmUlPBvAqTrSo0eDAkC1PzlJ2jdrko_QXCTHErjQUHKSygrK5zfCL0gk_BskvUgZOafdYcGbPNIV39MX6NDsINcbACVN6gdrVc-1u0776q-Wp5F2_2G7UzogE
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFH7oFNSLvyb-Ngev3ZombVKvurHhLGNM2EVGmqYykE62zr_fvKyb8-DBU0MJISQh7-W9930fwL0fKppi1EYzgdEqobxUoogAU9zPZa51kDmxCZEkcjSK-xVY3WFhjDGu-Mw0sOly-dlULzBU1qSIdwkY24adkNuHzxKutTo-mODjPl8nEeyzXkYVmw_lYRO5yUIWNfAboU76hpyKsybtw3_O4wjqP7g80l9bnGPYMsUJHGxQCp7C28DMHb7K6yaDB_IL_0Faji_Cjk2mhfdk8JIgFcHqO3HFA6TrKswnJUHWDvVBBq5StgIoFXV4bbeGjx2vklDwVBDT0tMiQ0Z1Y7j1jDQVGeU0Uzw1gfZlLnQUB1TFXMRBxIy0nVOZx-jD0NB2zXx2BrViWphzIEzyVKgUXSQ7gEzjZZYvsi3rVbDwAuq4UuPPJUvGeLVIl3_8v4O9zvClN-51k-cr2Le7xdEmBOIaauVsYW5gV3-Vk_ns1u3yN1PEpUg
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=Digest+of+technical+papers+-+IEEE%2FACM+International+Conference+on+Computer-Aided+Design&rft.atitle=Residual-INR%3A+Communication+Efficient+on-Device+Learning+Using+Implicit+Neural+Representation&rft.au=Chen%2C+Hanqiu&rft.au=Yao%2C+Xuebin&rft.au=Subedi%2C+Pradeep&rft.au=Hao%2C+Cong&rft.date=2024-10-27&rft.pub=ACM&rft.eissn=1558-2434&rft.spage=1&rft.epage=9&rft_id=info:doi/10.1145%2F3676536.3676673&rft.externalDocID=11126233