Sparse Generalized Eigenvalue Problem Via Smooth Optimization

In this paper, we consider an ℓ 0 -norm penalized formulation of the generalized eigenvalue problem (GEP), aimed at extracting the leading sparse generalized eigenvector of a matrix pair. The formulation involves maximization of a discontinuous nonconcave objective function over a nonconvex constrai...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on signal processing Ročník 63; číslo 7; s. 1627 - 1642
Hlavní autoři: Junxiao Song, Babu, Prabhu, Palomar, Daniel P.
Médium: Journal Article
Jazyk:angličtina
Vydáno: IEEE 01.04.2015
Témata:
ISSN:1053-587X, 1941-0476
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract In this paper, we consider an ℓ 0 -norm penalized formulation of the generalized eigenvalue problem (GEP), aimed at extracting the leading sparse generalized eigenvector of a matrix pair. The formulation involves maximization of a discontinuous nonconcave objective function over a nonconvex constraint set, and is therefore computationally intractable. To tackle the problem, we first approximate the ℓ 0 -norm by a continuous surrogate function. Then an algorithm is developed via iteratively majorizing the surrogate function by a quadratic separable function, which at each iteration reduces to a regular generalized eigenvalue problem. A preconditioned steepest ascent algorithm for finding the leading generalized eigenvector is provided. A systematic way based on smoothing is proposed to deal with the "singularity issue" that arises when a quadratic function is used to majorize the nondifferentiable surrogate function. For sparse GEPs with special structure, algorithms that admit a closed-form solution at every iteration are derived. Numerical experiments show that the proposed algorithms match or outperform existing algorithms in terms of computational complexity and support recovery.
AbstractList In this paper, we consider an ℓ 0 -norm penalized formulation of the generalized eigenvalue problem (GEP), aimed at extracting the leading sparse generalized eigenvector of a matrix pair. The formulation involves maximization of a discontinuous nonconcave objective function over a nonconvex constraint set, and is therefore computationally intractable. To tackle the problem, we first approximate the ℓ 0 -norm by a continuous surrogate function. Then an algorithm is developed via iteratively majorizing the surrogate function by a quadratic separable function, which at each iteration reduces to a regular generalized eigenvalue problem. A preconditioned steepest ascent algorithm for finding the leading generalized eigenvector is provided. A systematic way based on smoothing is proposed to deal with the "singularity issue" that arises when a quadratic function is used to majorize the nondifferentiable surrogate function. For sparse GEPs with special structure, algorithms that admit a closed-form solution at every iteration are derived. Numerical experiments show that the proposed algorithms match or outperform existing algorithms in terms of computational complexity and support recovery.
Author Junxiao Song
Babu, Prabhu
Palomar, Daniel P.
Author_xml – sequence: 1
  surname: Junxiao Song
  fullname: Junxiao Song
  email: jsong@ust.hk
  organization: Dept. of Electron. & Comput. Eng., Hong Kong Univ. of Sci. & Technol. (HKUST), Hong Kong, China
– sequence: 2
  givenname: Prabhu
  surname: Babu
  fullname: Babu, Prabhu
  email: eeprabhubabu@ust.hk
  organization: Dept. of Electron. & Comput. Eng., Hong Kong Univ. of Sci. & Technol. (HKUST), Hong Kong, China
– sequence: 3
  givenname: Daniel P.
  surname: Palomar
  fullname: Palomar, Daniel P.
  email: palomar@ust.hk
  organization: Dept. of Electron. & Comput. Eng., Hong Kong Univ. of Sci. & Technol. (HKUST), Hong Kong, China
BookMark eNp9kE1Lw0AQhhepYFu9C17yBxJ3sptN9uBBSq1CoYVW8RYm21ldyUfZRKH99aa2ePAgc3jn8D7D8IzYoG5qYuwaeATA9e16tYxiDkkUCy2lFGdsCFpCyGWqBv3OExEmWfp6wUZt-8E5SKnVkN2ttuhbCmZUk8fS7WkTTN0b1V9YflKw9E1RUhW8OAxWVdN078Fi27nK7bFzTX3Jzi2WLV2dcsyeH6bryWM4X8yeJvfz0MRKdKFKSaOQsZVKZbEqVCJRFwigMFOZsklmuJVkCiOAoADRD2yM3RBq25fFmKnjXeObtvVkc-O6nw86j67MgecHCXkvIT9IyE8SepD_AbfeVeh3_yE3R8QR0W895ZD2-sQ34CFpyA
CODEN ITPRED
CitedBy_id crossref_primary_10_1109_LCOMM_2024_3438876
crossref_primary_10_1109_LSP_2021_3108675
crossref_primary_10_1109_TSP_2018_2799193
crossref_primary_10_1109_LSP_2016_2593589
crossref_primary_10_1137_20M1325381
crossref_primary_10_1109_TVT_2019_2962542
crossref_primary_10_1109_TSP_2017_2762286
crossref_primary_10_1088_1361_6420_acbe5e
crossref_primary_10_1111_biom_12886
crossref_primary_10_1109_TVT_2023_3298223
crossref_primary_10_1016_j_sigpro_2019_107433
crossref_primary_10_1109_TCOMM_2018_2854175
crossref_primary_10_1080_10618600_2019_1568014
crossref_primary_10_1109_TSP_2021_3058442
crossref_primary_10_1109_LSP_2017_2698721
crossref_primary_10_1109_TSP_2015_2452219
crossref_primary_10_1016_j_sigpro_2018_12_017
crossref_primary_10_1016_j_sigpro_2022_108914
crossref_primary_10_1109_TSP_2016_2601299
crossref_primary_10_1016_j_conengprac_2019_07_007
crossref_primary_10_1007_s11222_025_10610_0
crossref_primary_10_1109_TNSE_2019_2952454
crossref_primary_10_1109_TSP_2016_2605073
crossref_primary_10_1109_TCYB_2020_2982901
crossref_primary_10_1109_LSP_2021_3132276
crossref_primary_10_1137_22M1472000
crossref_primary_10_1016_j_jfranklin_2020_03_034
Cites_doi 10.1137/050645506
10.1137/090756855
10.1038/nature04296
10.1109/TSP.2013.2293126
10.1007/s10994-010-5226-3
10.1137/120891009
10.1080/03610927708827533
10.1007/s10107-004-0552-5
10.1109/ICASSP.2008.4518498
10.1109/TSP.2009.2031732
10.1137/1.9780898719581
10.1080/757584614
10.1137/110839072
10.1007/s10898-004-3134-4
10.1287/moor.1070.0268
10.1109/78.558475
10.1080/01621459.1973.10481436
10.1198/106186006X113430
10.1093/biostatistics/kxp008
10.1109/TIP.2007.909318
10.1007/s00041-008-9045-x
10.2307/2333955
10.1007/978-3-642-99789-1_13
10.1198/0003130042836
10.1016/j.jmva.2007.06.007
10.1145/1273496.1273601
10.1016/j.compbiolchem.2005.08.006
10.1109/78.738251
ContentType Journal Article
DBID 97E
RIA
RIE
AAYXX
CITATION
DOI 10.1109/TSP.2015.2394443
DatabaseName IEEE Xplore (IEEE)
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 1941-0476
EndPage 1642
ExternalDocumentID 10_1109_TSP_2015_2394443
7017587
Genre orig-research
GrantInformation_xml – fundername: Hong Kong RGC 617312
GroupedDBID -~X
.DC
0R~
29I
4.4
5GY
6IK
85S
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACIWK
ACNCT
AENEX
AGQYO
AGSQL
AHBIQ
AJQPL
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ASUFR
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
EBS
EJD
F5P
HZ~
IFIPE
IPLJI
JAVBF
LAI
MS~
O9-
OCL
P2P
RIA
RIE
RNS
TAE
TN5
3EH
53G
5VS
AAYXX
ABFSI
ACKIV
AETIX
AI.
AIBXA
AKJIK
ALLEH
CITATION
E.L
H~9
ICLAB
IFJZH
VH1
ID FETCH-LOGICAL-c263t-67e9a342f466826b654a9ba116a8686f58c0f4ecbc31e1b131311dcfdea9f6543
IEDL.DBID RIE
ISICitedReferencesCount 53
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000350880900001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1053-587X
IngestDate Tue Nov 18 22:32:10 EST 2025
Sat Nov 29 04:10:34 EST 2025
Tue Aug 26 16:40:39 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 7
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c263t-67e9a342f466826b654a9ba116a8686f58c0f4ecbc31e1b131311dcfdea9f6543
PageCount 16
ParticipantIDs ieee_primary_7017587
crossref_citationtrail_10_1109_TSP_2015_2394443
crossref_primary_10_1109_TSP_2015_2394443
PublicationCentury 2000
PublicationDate 2015-April1,
2015-4-00
PublicationDateYYYYMMDD 2015-04-01
PublicationDate_xml – month: 04
  year: 2015
  text: 2015-April1,
  day: 01
PublicationDecade 2010
PublicationTitle IEEE transactions on signal processing
PublicationTitleAbbrev TSP
PublicationYear 2015
Publisher IEEE
Publisher_xml – name: IEEE
References ref13
ref12
ref15
ref14
ref31
ref30
ref11
ref10
ref2
ref17
ref16
ref19
ref18
jolliffe (ref1) 2002
journ e (ref7) 2010; 11
ref24
ref23
ref26
ref25
ref20
ref22
ref21
ref28
rockafellar (ref27) 1997; 28
ref29
ref8
ref9
ref4
ref3
ref6
ref5
References_xml – ident: ref5
  doi: 10.1137/050645506
– ident: ref24
  doi: 10.1137/090756855
– ident: ref29
  doi: 10.1038/nature04296
– ident: ref21
  doi: 10.1109/TSP.2013.2293126
– ident: ref11
  doi: 10.1007/s10994-010-5226-3
– ident: ref20
  doi: 10.1137/120891009
– ident: ref22
  doi: 10.1080/03610927708827533
– ident: ref14
  doi: 10.1007/s10107-004-0552-5
– ident: ref16
  doi: 10.1109/ICASSP.2008.4518498
– year: 2002
  ident: ref1
  publication-title: Principal Component Analysis
– volume: 11
  start-page: 517
  year: 2010
  ident: ref7
  article-title: Generalized power method for sparse principal component analysis
  publication-title: J Mach Learn Res
– volume: 28
  year: 1997
  ident: ref27
  publication-title: Convex Analysis
– ident: ref31
  doi: 10.1109/TSP.2009.2031732
– ident: ref25
  doi: 10.1137/1.9780898719581
– ident: ref3
  doi: 10.1080/757584614
– ident: ref8
  doi: 10.1137/110839072
– ident: ref26
  doi: 10.1007/s10898-004-3134-4
– ident: ref30
  doi: 10.1287/moor.1070.0268
– ident: ref15
  doi: 10.1109/78.558475
– ident: ref12
  doi: 10.1080/01621459.1973.10481436
– ident: ref4
  doi: 10.1198/106186006X113430
– ident: ref9
  doi: 10.1093/biostatistics/kxp008
– ident: ref13
  doi: 10.1109/TIP.2007.909318
– ident: ref17
  doi: 10.1007/s00041-008-9045-x
– ident: ref2
  doi: 10.2307/2333955
– ident: ref18
  doi: 10.1007/978-3-642-99789-1_13
– ident: ref19
  doi: 10.1198/0003130042836
– ident: ref6
  doi: 10.1016/j.jmva.2007.06.007
– ident: ref10
  doi: 10.1145/1273496.1273601
– ident: ref28
  doi: 10.1016/j.compbiolchem.2005.08.006
– ident: ref23
  doi: 10.1109/78.738251
SSID ssj0014496
Score 2.4056957
Snippet In this paper, we consider an ℓ 0 -norm penalized formulation of the generalized eigenvalue problem (GEP), aimed at extracting the leading sparse generalized...
SourceID crossref
ieee
SourceType Enrichment Source
Index Database
Publisher
StartPage 1627
SubjectTerms Approximation methods
Eigenvalues and eigenfunctions
Minorization-maximization
Principal component analysis
Signal processing algorithms
smooth optimization
sparse generalized eigenvalue problem
Sparse matrices
sparse PCA
Tin
Vectors
Title Sparse Generalized Eigenvalue Problem Via Smooth Optimization
URI https://ieeexplore.ieee.org/document/7017587
Volume 63
WOSCitedRecordID wos000350880900001&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 Electronic Library (IEL)
  customDbUrl:
  eissn: 1941-0476
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0014496
  issn: 1053-587X
  databaseCode: RIE
  dateStart: 19910101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEB5q8aAHX1WsL_bgRTDtbpNNNkeRFg9SC63S25JNJlCwrfThwV9vsrtdKojgbVlmYPkSMjM7me8DuLWZcIeepsS4cEWYoIK4sjkmFKUWGU0ML9j1n0W_n4zHclCD-2oWBhHzy2fY8o95L9_M9dr_KmsLt33iROzAjhC8mNWqOgaM5VpcLl2gxNmMNy3JULZHw4G_wxW3vAw4Y_RHCNrSVMlDSu_wfx9zBAdl6hg8FGt9DDWcncD-FqFgA7zC8GKJQUkmPflCE3Q936bn9MZgUKjHBG8TFQync7dIwYs7MqblLOYpvPa6o8cnUgokEN3hdEW4QKko61jGuSsTMh4zJTMVRVwlPOE2TnRoGepM0wijLKKeW8doa1BJ64dKz6A-m8_wHAIupNAdExp0Hkr6yTujXKoVW86tSyGa0N5gluqSPdyLWLyneRURytShnHqU0xLlJtxVHh8Fc8Yftg0PcGVXYnvx--tL2PPOxf2ZK6ivFmu8hl39uZosFzf5tvgGIaaz2Q
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1bS8MwFD7MKagP3qY4r3nwRbBbu6RJ8yiyMXHOwabsraRJCgO3yS4--OtN2qxMEMG3Uk5C-RJyzmnO-T6AmzRh5tCT2FPGXXmEYeaZtDn0sOaSJThSNGfX77BuNxoOea8Ed0UvjNY6Kz7TNfuY3eWrqVzaX2V1ZrZPGLEN2LTKWa5bq7gzICRT4zIBA_aM1XB1Kenz-qDfs1VcYc0KgROCfzihNVWVzKm09v_3OQew54JHdJ-v9iGU9OQIdtcoBStgNYZnc40cnfToSyvUtIybltVbo16uH4PeRgL1x1OzTOjFHBpj1415DK-t5uCh7TmJBE82KF54lGkuMGmkhFKTKCQ0JIInIgioiGhE0zCSfkq0TCQOdJAE2LLrKJkqLXhq20pPoDyZTvQpIMo4kw3lK21GCG5775QwwVaYUpqaIKIK9RVmsXT84VbG4j3O8gifxwbl2KIcO5SrcFuM-Mi5M_6wrViACzuH7dnvr69huz147sSdx-7TOezYifJqmgsoL2ZLfQlb8nMxms-usi3yDbADtyI
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=Sparse+Generalized+Eigenvalue+Problem+Via+Smooth+Optimization&rft.jtitle=IEEE+transactions+on+signal+processing&rft.au=Junxiao+Song&rft.au=Babu%2C+Prabhu&rft.au=Palomar%2C+Daniel+P.&rft.date=2015-04-01&rft.pub=IEEE&rft.issn=1053-587X&rft.volume=63&rft.issue=7&rft.spage=1627&rft.epage=1642&rft_id=info:doi/10.1109%2FTSP.2015.2394443&rft.externalDocID=7017587
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1053-587X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1053-587X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1053-587X&client=summon