Robust Low-Tubal-Rank Tensor Completion Based on Tensor Factorization and Maximum Correntopy Criterion

The goal of tensor completion is to recover a tensor from a subset of its entries, often by exploiting its low-rank property. Among several useful definitions of tensor rank, the low tubal rank was shown to give a valuable characterization of the inherent low-rank structure of a tensor. While some l...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE transaction on neural networks and learning systems Ročník 35; číslo 10; s. 14603 - 14617
Hlavní autori: He, Yicong, Atia, George K.
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States IEEE 01.10.2024
Predmet:
ISSN:2162-237X, 2162-2388, 2162-2388
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract The goal of tensor completion is to recover a tensor from a subset of its entries, often by exploiting its low-rank property. Among several useful definitions of tensor rank, the low tubal rank was shown to give a valuable characterization of the inherent low-rank structure of a tensor. While some low-tubal-rank tensor completion algorithms with favorable performance have been recently proposed, these algorithms utilize second-order statistics to measure the error residual, which may not work well when the observed entries contain large outliers. In this article, we propose a new objective function for low-tubal-rank tensor completion, which uses correntropy as the error measure to mitigate the effect of the outliers. To efficiently optimize the proposed objective, we leverage a half-quadratic minimization technique whereby the optimization is transformed to a weighted low-tubal-rank tensor factorization problem. Subsequently, we propose two simple and efficient algorithms to obtain the solution and provide their convergence and complexity analysis. Numerical results using both synthetic and real data demonstrate the robust and superior performance of the proposed algorithms.
AbstractList The goal of tensor completion is to recover a tensor from a subset of its entries, often by exploiting its low-rank property. Among several useful definitions of tensor rank, the low tubal rank was shown to give a valuable characterization of the inherent low-rank structure of a tensor. While some low-tubal-rank tensor completion algorithms with favorable performance have been recently proposed, these algorithms utilize second-order statistics to measure the error residual, which may not work well when the observed entries contain large outliers. In this article, we propose a new objective function for low-tubal-rank tensor completion, which uses correntropy as the error measure to mitigate the effect of the outliers. To efficiently optimize the proposed objective, we leverage a half-quadratic minimization technique whereby the optimization is transformed to a weighted low-tubal-rank tensor factorization problem. Subsequently, we propose two simple and efficient algorithms to obtain the solution and provide their convergence and complexity analysis. Numerical results using both synthetic and real data demonstrate the robust and superior performance of the proposed algorithms.
The goal of tensor completion is to recover a tensor from a subset of its entries, often by exploiting its low-rank property. Among several useful definitions of tensor rank, the low tubal rank was shown to give a valuable characterization of the inherent low-rank structure of a tensor. While some low-tubal-rank tensor completion algorithms with favorable performance have been recently proposed, these algorithms utilize second-order statistics to measure the error residual, which may not work well when the observed entries contain large outliers. In this article, we propose a new objective function for low-tubal-rank tensor completion, which uses correntropy as the error measure to mitigate the effect of the outliers. To efficiently optimize the proposed objective, we leverage a half-quadratic minimization technique whereby the optimization is transformed to a weighted low-tubal-rank tensor factorization problem. Subsequently, we propose two simple and efficient algorithms to obtain the solution and provide their convergence and complexity analysis. Numerical results using both synthetic and real data demonstrate the robust and superior performance of the proposed algorithms.The goal of tensor completion is to recover a tensor from a subset of its entries, often by exploiting its low-rank property. Among several useful definitions of tensor rank, the low tubal rank was shown to give a valuable characterization of the inherent low-rank structure of a tensor. While some low-tubal-rank tensor completion algorithms with favorable performance have been recently proposed, these algorithms utilize second-order statistics to measure the error residual, which may not work well when the observed entries contain large outliers. In this article, we propose a new objective function for low-tubal-rank tensor completion, which uses correntropy as the error measure to mitigate the effect of the outliers. To efficiently optimize the proposed objective, we leverage a half-quadratic minimization technique whereby the optimization is transformed to a weighted low-tubal-rank tensor factorization problem. Subsequently, we propose two simple and efficient algorithms to obtain the solution and provide their convergence and complexity analysis. Numerical results using both synthetic and real data demonstrate the robust and superior performance of the proposed algorithms.
Author Atia, George K.
He, Yicong
Author_xml – sequence: 1
  givenname: Yicong
  orcidid: 0000-0003-3398-3376
  surname: He
  fullname: He, Yicong
  email: yicong.he@ucf.edu
  organization: Department of Electrical and Computer Engineering, University of Central Florida, Orlando, FL, USA
– sequence: 2
  givenname: George K.
  orcidid: 0000-0001-7958-9855
  surname: Atia
  fullname: Atia, George K.
  email: george.atia@ucf.edu
  organization: Department of Electrical and Computer Engineering and the Department of Computer Science, University of Central Florida, Orlando, FL, USA
BackLink https://www.ncbi.nlm.nih.gov/pubmed/37279124$$D View this record in MEDLINE/PubMed
BookMark eNp9kUtLxDAUhYMovv-AiHTppmN606bNUgdfMCroCO5Kmt5CtG3GJMXHrzfzUMSF2eTA-U4uuWeHrPemR0IOEjpKEipOpre3k4cRUGAjBgWlBV8j25BwiIEVxfqPzp-2yL5zzzQcTjOeik2yxXLIRQLpNmnuTTU4H03MWzwdKtnG97J_iabYO2OjselmLXpt-uhMOqyjIFbWhVTeWP0pF67s6-hGvutu6ELIWuy9mX1EY6s92gDskY1Gtg73V_cuebw4n46v4snd5fX4dBIrBqmPc6HqSvEUc84qLptMSORNLUAwRYMEqhrKa0DMsvAHoRTNoAaWsURxJgXbJcfLd2fWvA7ofNlpp7BtZY9mcCUUwFKRc5ijRyt0qDqsy5nVnbQf5fduAgBLQFnjnMXmB0loOe-gXHRQzjsoVx2EUPEnpLRf7Mhbqdv_o4fLqEbEX7OSNOXB_gISaZTx
CODEN ITNNAL
CitedBy_id crossref_primary_10_1049_ipr2_70171
crossref_primary_10_1109_LSP_2025_3597829
crossref_primary_10_1109_TASE_2025_3595545
crossref_primary_10_1016_j_sigpro_2025_109935
Cites_doi 10.24963/ijcai.2019/368
10.1080/03610917808812083
10.1137/030600862
10.1109/TCI.2020.3006718
10.1109/IJCNN.2009.5178823
10.1002/1099-128X(200005/06)14:3<105::AID-CEM582>3.0.CO;2-I
10.1109/TNNLS.2016.2611525
10.1109/CVPR.2016.85
10.1109/TKDE.2019.2937027
10.1109/ICCV.2009.5459186
10.1109/TNNLS.2015.2465178
10.1007/s11263-018-1086-2
10.1109/TSP.2007.896065
10.1109/TNNLS.2020.2979685
10.1109/TSP.2019.2952057
10.1137/110837711
10.1109/TNNLS.2018.2796606
10.1016/j.automatica.2016.10.004
10.1007/s10208-009-9045-5
10.1109/TSP.2017.2690524
10.1016/j.acha.2015.08.003
10.1109/MSP.2013.2297439
10.1016/j.laa.2010.09.020
10.1137/130905010
10.1109/TPAMI.2019.2891760
10.3934/ipi.2015.9.601
10.1007/BF02289464
10.1109/ICASSP.2019.8683818
10.1109/ICCV.2019.01028
10.1109/TIP.2017.2762595
10.1109/TSP.2016.2539127
10.1109/JPROC.2009.2035722
10.1109/LSP.2009.2018223
10.1109/IJCNN.2011.6033473
10.1016/j.patcog.2013.07.017
10.1109/CVPR.2019.01240
10.1109/TPAMI.2010.220
10.1109/TNNLS.2020.3009417
10.1109/CVPR.2017.411
10.1109/TCYB.2021.3053057
10.1109/TSP.2016.2639466
10.1109/TIT.2019.2959980
ContentType Journal Article
DBID 97E
RIA
RIE
AAYXX
CITATION
NPM
7X8
DOI 10.1109/TNNLS.2023.3280086
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE/IET Electronic Library
CrossRef
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
PubMed
MEDLINE - Academic
DatabaseTitleList PubMed
MEDLINE - Academic

Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: RIE
  name: IEEE/IET Electronic Library
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 2162-2388
EndPage 14617
ExternalDocumentID 37279124
10_1109_TNNLS_2023_3280086
10144686
Genre orig-research
Journal Article
GrantInformation_xml – fundername: NSF Award
  grantid: CCF-2106339
– fundername: NSF CAREER Award
  grantid: CCF-1552497
  funderid: 10.13039/100000001
GroupedDBID 0R~
4.4
5VS
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACIWK
ACPRK
AENEX
AFRAH
AGQYO
AGSQL
AHBIQ
AKJIK
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
EBS
EJD
IFIPE
IPLJI
JAVBF
M43
MS~
O9-
OCL
PQQKQ
RIA
RIE
RNS
AAYXX
CITATION
NPM
RIG
7X8
ID FETCH-LOGICAL-c324t-79cdbc64e763b6af59ae6fd9293c0ae620cf06d2ee553729cc052d23531c63a93
IEDL.DBID RIE
ISICitedReferencesCount 6
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001006026100001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2162-237X
2162-2388
IngestDate Sat Sep 27 18:59:40 EDT 2025
Mon Jul 21 05:49:04 EDT 2025
Sat Nov 29 01:40:26 EST 2025
Tue Nov 18 20:53:10 EST 2025
Wed Aug 27 02:19:18 EDT 2025
IsPeerReviewed false
IsScholarly true
Issue 10
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
https://doi.org/10.15223/policy-029
https://doi.org/10.15223/policy-037
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c324t-79cdbc64e763b6af59ae6fd9293c0ae620cf06d2ee553729cc052d23531c63a93
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0003-3398-3376
0000-0001-7958-9855
PMID 37279124
PQID 2823497629
PQPubID 23479
PageCount 15
ParticipantIDs proquest_miscellaneous_2823497629
crossref_primary_10_1109_TNNLS_2023_3280086
pubmed_primary_37279124
crossref_citationtrail_10_1109_TNNLS_2023_3280086
ieee_primary_10144686
PublicationCentury 2000
PublicationDate 2024-10-01
PublicationDateYYYYMMDD 2024-10-01
PublicationDate_xml – month: 10
  year: 2024
  text: 2024-10-01
  day: 01
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle IEEE transaction on neural networks and learning systems
PublicationTitleAbbrev TNNLS
PublicationTitleAlternate IEEE Trans Neural Netw Learn Syst
PublicationYear 2024
Publisher IEEE
Publisher_xml – name: IEEE
References ref12
ref15
Gilman (ref23) 2020
ref14
ref11
ref10
ref16
ref19
ref46
ref45
ref48
ref47
ref42
Kasai (ref18)
ref41
ref44
ref43
ref8
ref7
ref9
ref4
ref3
ref6
He (ref5)
ref40
ref35
ref34
ref37
ref36
ref31
ref30
ref33
ref32
ref2
ref1
ref39
ref38
Huang (ref27) 2014; 4252
ref24
ref26
Zhao (ref13) 2016
ref25
ref20
ref22
ref21
ref28
ref29
Jain (ref17)
Mallick (ref49) 2020
References_xml – ident: ref31
  doi: 10.24963/ijcai.2019/368
– ident: ref46
  doi: 10.1080/03610917808812083
– ident: ref40
  doi: 10.1137/030600862
– ident: ref29
  doi: 10.1109/TCI.2020.3006718
– ident: ref44
  doi: 10.1109/IJCNN.2009.5178823
– start-page: 1012
  volume-title: Proc. Int. Conf. Mach. Learn.
  ident: ref18
  article-title: Low-rank tensor completion: A Riemannian manifold preconditioning approach
– ident: ref11
  doi: 10.1002/1099-128X(200005/06)14:3<105::AID-CEM582>3.0.CO;2-I
– ident: ref42
  doi: 10.1109/TNNLS.2016.2611525
– ident: ref48
  doi: 10.1109/CVPR.2016.85
– volume: 4252
  start-page: 455
  issue: 2
  year: 2014
  ident: ref27
  article-title: Provable low-rank tensor recovery
  publication-title: Optim.-Online
– ident: ref10
  doi: 10.1109/TKDE.2019.2937027
– ident: ref26
  doi: 10.1109/ICCV.2009.5459186
– ident: ref28
  doi: 10.1109/TNNLS.2015.2465178
– ident: ref3
  doi: 10.1007/s11263-018-1086-2
– ident: ref33
  doi: 10.1109/TSP.2007.896065
– ident: ref6
  doi: 10.1109/TNNLS.2020.2979685
– ident: ref36
  doi: 10.1109/TSP.2019.2952057
– ident: ref30
  doi: 10.1137/110837711
– ident: ref25
  doi: 10.1109/TNNLS.2018.2796606
– ident: ref38
  doi: 10.1016/j.automatica.2016.10.004
– start-page: 499
  volume-title: Proc. Adv. Neural Inf. Process. Syst.
  ident: ref5
  article-title: Tensor subspace analysis
– ident: ref15
  doi: 10.1007/s10208-009-9045-5
– ident: ref8
  doi: 10.1109/TSP.2017.2690524
– ident: ref47
  doi: 10.1016/j.acha.2015.08.003
– ident: ref9
  doi: 10.1109/MSP.2013.2297439
– ident: ref14
  doi: 10.1016/j.laa.2010.09.020
– ident: ref24
  doi: 10.1137/130905010
– ident: ref41
  doi: 10.1109/TPAMI.2019.2891760
– ident: ref19
  doi: 10.3934/ipi.2015.9.601
– ident: ref12
  doi: 10.1007/BF02289464
– ident: ref32
  doi: 10.1109/ICASSP.2019.8683818
– ident: ref2
  doi: 10.1109/ICCV.2019.01028
– year: 2016
  ident: ref13
  article-title: Tensor ring decomposition
  publication-title: arXiv:1606.05535
– ident: ref21
  doi: 10.1109/TIP.2017.2762595
– ident: ref34
  doi: 10.1109/TSP.2016.2539127
– ident: ref16
  doi: 10.1109/JPROC.2009.2035722
– start-page: 1431
  volume-title: Proc. Adv. Neural Inf. Process. Syst.
  ident: ref17
  article-title: Provable tensor factorization with missing data
– ident: ref43
  doi: 10.1109/LSP.2009.2018223
– ident: ref39
  doi: 10.1109/IJCNN.2011.6033473
– year: 2020
  ident: ref23
  article-title: Grassmannian optimization for online tensor completion and tracking with the t-SVD
  publication-title: arXiv:2001.11419
– ident: ref45
  doi: 10.1016/j.patcog.2013.07.017
– ident: ref4
  doi: 10.1109/CVPR.2019.01240
– ident: ref35
  doi: 10.1109/TPAMI.2010.220
– ident: ref37
  doi: 10.1109/TNNLS.2020.3009417
– year: 2020
  ident: ref49
  article-title: Transfer learning with graph neural networks for short-term highway traffic forecasting
  publication-title: arXiv:2004.08038
– ident: ref1
  doi: 10.1109/CVPR.2017.411
– ident: ref7
  doi: 10.1109/TCYB.2021.3053057
– ident: ref20
  doi: 10.1109/TSP.2016.2639466
– ident: ref22
  doi: 10.1109/TIT.2019.2959980
SSID ssj0000605649
Score 2.5124464
Snippet The goal of tensor completion is to recover a tensor from a subset of its entries, often by exploiting its low-rank property. Among several useful definitions...
SourceID proquest
pubmed
crossref
ieee
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 14603
SubjectTerms Alternating minimization
Computational modeling
correntropy
half-quadratic (HQ)
Linear programming
Matrix decomposition
Measurement uncertainty
Minimization
Signal processing algorithms
tensor completion
tensor factorization
Tensors
Title Robust Low-Tubal-Rank Tensor Completion Based on Tensor Factorization and Maximum Correntopy Criterion
URI https://ieeexplore.ieee.org/document/10144686
https://www.ncbi.nlm.nih.gov/pubmed/37279124
https://www.proquest.com/docview/2823497629
Volume 35
WOSCitedRecordID wos001006026100001&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: 2162-2388
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000605649
  issn: 2162-237X
  databaseCode: RIE
  dateStart: 20120101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dT9swED9BhSZeVrYx6PiQJ_GGUlI7ceLHgaj2wKoJitS3yF-REBCjthnw33N2kgoemMSbpZyTKHeX-53tux_AkYi5QmAxikya-QQlQ58rFY0weS6tZElKlQxkE9lkks9m4m9brB5qYay14fCZHfph2Ms3Ttd-qezE88omPOfrsJ5lvCnWWi2oxAjMeYC7dMRpRFk264pkYnEynUwuroaeK3zIaO6B_CZ8Yhi8xYgmb2JSIFl5H2-GuDPuf_CNt-BzCzDJr8YivsCarb5CvyNvIK0vf4Py0ql6sSQX7jGa1kreRZeyuiVTTGvdnHh535bbVeQU45whOGgvjQNDT1u-SWRlyB_5dHNf3-Ok0OvJPTwTz6CAtu2qbbgen0_Pfkct6UKkEVsto0xoozRPLP54FJdlKqTlpUEUxXSMQxrrMuaGWpumfstP6zilhjL0Zc2ZFOw79CpX2V0gXEstQn99jgl4mSupUFrrNDUMfxXJAEbdZy9025HcE2PcFSEziUURtFZ4rRWt1gZwvJrz0PTj-K_0ttfJK8lGHQP42am3QHfyeySysq5eFJiBsgQhGhUD2Gn0vprdmcuPd-66B5v48KQ56rcPveW8tgewof8tbxbzQ7TZWX4YbPYFiunlLA
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Lb9QwEB5BQdAL5VFgeRqJG8rW61fWR0CsithGqARpb5HtOFJFG1e7Gx7_nrGTrOBQJG6WMo6szIznm9gzH8BrTZVFYDHLapnHBCVHn2ssyzB5brzhQjJrEtlEXhTz1Up_HorVUy2M9z5dPvPTOExn-XVwXfxVdhR5ZYWaq-twQwrBaF-utfulQhGaqwR42UyxjPF8NZbJUH1UFsXyyzSyhU85m0covw-3OIZvPWPir6iUaFauRpwp8iwO_nPNd-HOADHJ294m7sE1396Hg5G-gQze_ACa02C7zZYsw4-s7Kw5z05N-42UmNiGNYnysTF3aMk7jHQ1wcHwaJE4eoYCTmLampyYn2cX3QVOSt2ewuUvEjkU0LpDewhfFx_K98fZQLuQOURX2yzXrrZOCY9bj1Wmkdp41dSIo7ijOGTUNVTVzHsp46Gfc1SymnH0Zqe40fwh7LWh9Y-BKGecTh32Fabgzdwai9LOSVlz3CzEBGbjZ6_c0JM8UmOcVyk3obpKWqui1qpBaxN4s5tz2Xfk-Kf0YdTJH5K9OibwalRvhQ4VT0lM60O3qTAH5QJBGtMTeNTrfTd7NJcnV7z1Jdw-Lk-W1fJj8ekp7ONCRH_x7xnsbdedfw433fft2Wb9IlnubzX554s
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=Robust+Low-Tubal-Rank+Tensor+Completion+Based+on+Tensor+Factorization+and+Maximum+Correntopy+Criterion&rft.jtitle=IEEE+transaction+on+neural+networks+and+learning+systems&rft.au=He%2C+Yicong&rft.au=Atia%2C+George+K&rft.date=2024-10-01&rft.eissn=2162-2388&rft.volume=35&rft.issue=10&rft.spage=14603&rft_id=info:doi/10.1109%2FTNNLS.2023.3280086&rft_id=info%3Apmid%2F37279124&rft.externalDocID=37279124
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2162-237X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2162-237X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2162-237X&client=summon