A parallel decomposition algorithm for training multiclass kernel-based vector machines

We present a decomposition method for training Crammer and Singer's multiclass kernel-based vector machine model. A new working set selection rule is proposed. Global convergence of the algorithm based on this selection rule is established. Projected gradient method is chosen to solve the resul...

Full description

Saved in:
Bibliographic Details
Published inOptimization methods & software Vol. 26; no. 3; pp. 431 - 454
Main Authors Niu, Lingfeng, Yuan, Ya-Xiang
Format Journal Article
LanguageEnglish
Published Abingdon Taylor & Francis 01.06.2011
Taylor & Francis Ltd
Subjects
Algorithms
decomposition method
Kernel-based vector machines
parallel algorithm
projected gradient
Studies
Online AccessGet full text

Cover

Abstract We present a decomposition method for training Crammer and Singer's multiclass kernel-based vector machine model. A new working set selection rule is proposed. Global convergence of the algorithm based on this selection rule is established. Projected gradient method is chosen to solve the resulting quadratic subproblem at each iteration. An efficient projection algorithm is designed by exploiting the structure of the constraints. Parallel strategies are given to utilize the storage and computational resources available on the multiprocessor system. Numerical experiment on benchmark problems demonstrates that the good classification accuracy and remarkable time saving can be achieved.
AbstractList We present a decomposition method for training Crammer and Singer's multiclass kernel-based vector machine model. A new working set selection rule is proposed. Global convergence of the algorithm based on this selection rule is established. Projected gradient method is chosen to solve the resulting quadratic subproblem at each iteration. An efficient projection algorithm is designed by exploiting the structure of the constraints. Parallel strategies are given to utilize the storage and computational resources available on the multiprocessor system. Numerical experiment on benchmark problems demonstrates that the good classification accuracy and remarkable time saving can be achieved.
Author Yuan, Ya-Xiang
Niu, Lingfeng
Author_xml – sequence: 1
  givenname: Lingfeng
  surname: Niu
  fullname: Niu, Lingfeng
  organization: State Key Laboratory of Scientific and Engineering Computing , Institute of Computational Mathematics and Scientific/Engineering Computing, AMSS, CAS
– sequence: 2
  givenname: Ya-Xiang
  surname: Yuan
  fullname: Yuan, Ya-Xiang
  email: yyx@lsec.cc.ac.cn
  organization: State Key Laboratory of Scientific and Engineering Computing , Institute of Computational Mathematics and Scientific/Engineering Computing, AMSS, CAS
BookMark eNqFkL1OwzAURi0EEm3hDRgi9hQ7thOXBVUVf1IllkqMluPYrYtjB9sF9e1JFFgYYLp3-M53dc8UnDrvFABXCM4RZPAGQUrLirF5ARGa93uJ8QmYIFgscrLA1emwU5oPmXMwjXEPISSIlBPwusw6EYS1ymaNkr7tfDTJeJcJu_XBpF2baR-yFIRxxm2z9mCTkVbEmL2p4JTNaxFVk30omfpcK-TOOBUvwJkWNqrL7zkDm4f7zeopX788Pq-W61xiTFKOaKPrmlWqpppgSLGSoq4YREzVsNC6WDQNhZWGEGFNNCH9h3WphWINUpjhGbgea7vg3w8qJr73h-D6i5wxXGBGEepDZAzJ4GMMSvMumFaEI0eQDwL5j0A-COSjwB67_YVJk8TgZpBh_4PvRti4Xl8rPn2wDU_iaH3QQThpIsd_NnwBUCuMSQ
CitedBy_id crossref_primary_10_1016_j_neucom_2018_03_069
crossref_primary_10_1016_j_neunet_2012_05_011
Cites_doi 10.1109/TNN.2002.1000139
10.1109/ICPR.1994.576879
10.1145/1961189.1961199
10.1145/130385.130401
10.1613/jair.105
10.1007/s10107-005-0595-2
10.1007/978-3-642-76153-9_5
10.1007/BF01585748
10.1007/978-1-4757-2440-0
10.1145/1281192.1281270
10.1162/153244302320884605
10.1017/CBO9780511809682
10.1023/A:1008663629662
10.1162/15324430260185628
10.1016/S0167-8191(03)00021-8
10.1137/S1052623497330963
10.1162/15324430152733142
10.1162/15324430152733133
10.7551/mitpress/1130.003.0015
10.1080/10556789908805759
10.1098/rsta.1909.0016
10.1145/641876.641880
10.1145/1401890.1401942
10.1023/A:1013637720281
10.1007/0-387-30065-1_4
ContentType Journal Article
Copyright Copyright Taylor & Francis Group, LLC 2011
Copyright Taylor & Francis Ltd. Jun 2011
Copyright_xml – notice: Copyright Taylor & Francis Group, LLC 2011
– notice: Copyright Taylor & Francis Ltd. Jun 2011
DBID AAYXX
CITATION
7SC
8FD
JQ2
L7M
L~C
L~D
DOI 10.1080/10556788.2011.556633
DatabaseName CrossRef
Computer and Information Systems Abstracts
Technology Research Database
ProQuest Computer Science Collection
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts – Academic
Computer and Information Systems Abstracts Professional
DatabaseTitle CrossRef
Computer and Information Systems Abstracts
Technology Research Database
Computer and Information Systems Abstracts – Academic
Advanced Technologies Database with Aerospace
ProQuest Computer Science Collection
Computer and Information Systems Abstracts Professional
DatabaseTitleList Computer and Information Systems Abstracts
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 1029-4937
EndPage 454
ExternalDocumentID 2424104771
10_1080_10556788_2011_556633
556633
GroupedDBID .4S
.7F
.DC
.QJ
0BK
0R~
123
29N
30N
4.4
5VS
AAENE
AAJMT
AALDU
AAMIU
AAPUL
AAQRR
ABCCY
ABDBF
ABFIM
ABHAV
ABJNI
ABLIJ
ABPAQ
ABPEM
ABTAI
ABXUL
ABXYU
ACGEJ
ACGFS
ACIWK
ACTIO
ACUHS
ADCVX
ADGTB
ADXPE
AEISY
AENEX
AEOZL
AEPSL
AEYOC
AFKVX
AGCQS
AGDLA
AGMYJ
AHDZW
AIJEM
AJWEG
AKBVH
AKOOK
ALMA_UNASSIGNED_HOLDINGS
ALQZU
AQRUH
ARCSS
AVBZW
AWYRJ
BLEHA
CAG
CCCUG
CE4
COF
CS3
DGEBU
DKSSO
DU5
EAP
EBS
EDO
EMK
EPL
EST
ESX
E~A
E~B
F5P
GTTXZ
H13
HF~
HZ~
H~P
I-F
IPNFZ
J.P
KYCEM
M4Z
NA5
NY~
O9-
P2P
PQQKQ
RIG
RNANH
ROSJB
RTWRZ
S-T
SNACF
TBQAZ
TDBHL
TEJ
TFL
TFT
TFW
TTHFI
TUROJ
TUS
TWF
UT5
UU3
ZGOLN
~S~
07G
1TA
AAGDL
AAHIA
AAIKQ
AAKBW
AAYXX
ACAGQ
ACGEE
ADYSH
AEUMN
AFRVT
AGLEN
AGROQ
AHMOU
AIYEW
ALCKM
AMEWO
AMPGV
AMVHM
AMXXU
BCCOT
BPLKW
C06
CITATION
CRFIH
DMQIW
DWIFK
EJD
IVXBP
LJTGL
NUSFT
QCRFL
TAQ
TFMCV
TOXWX
UB9
UU8
V3K
V4Q
7SC
8FD
ACTCW
JQ2
L7M
L~C
L~D
TASJS
ID FETCH-LOGICAL-c334t-15dfbb87eb5f43053ecab78018eb02ff29dd507f0013f4f44201b6fae8d1e383
ISSN 1055-6788
IngestDate Wed Aug 13 11:11:27 EDT 2025
Thu Apr 24 23:07:18 EDT 2025
Tue Jul 01 05:30:31 EDT 2025
Wed Dec 25 09:03:24 EST 2024
IsPeerReviewed true
IsScholarly true
Issue 3
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c334t-15dfbb87eb5f43053ecab78018eb02ff29dd507f0013f4f44201b6fae8d1e383
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
PQID 883238511
PQPubID 186278
PageCount 24
ParticipantIDs crossref_primary_10_1080_10556788_2011_556633
informaworld_taylorfrancis_310_1080_10556788_2011_556633
crossref_citationtrail_10_1080_10556788_2011_556633
proquest_journals_883238511
ProviderPackageCode CITATION
AAYXX
PublicationCentury 2000
PublicationDate 2011-06-00
PublicationDateYYYYMMDD 2011-06-01
PublicationDate_xml – month: 06
  year: 2011
  text: 2011-06-00
PublicationDecade 2010
PublicationPlace Abingdon
PublicationPlace_xml – name: Abingdon
PublicationTitle Optimization methods & software
PublicationYear 2011
Publisher Taylor & Francis
Taylor & Francis Ltd
Publisher_xml – name: Taylor & Francis
– name: Taylor & Francis Ltd
References CIT0030
CIT0032
CIT0031
CIT0033
Kreßl U. (CIT0028) 1999
Sun W. (CIT0041) 2006
CIT0036
CIT0035
Chang C.-C. (CIT0008)
CIT0037
CIT0040
CIT0042
CIT0001
CIT0044
CIT0003
CIT0047
CIT0002
CIT0005
CIT0004
CIT0048
CIT0007
CIT0006
Tsochantaridis I. (CIT0043) 2005; 6
CIT0010
Vapnik V. N. (CIT0046) 1998
CIT0012
CIT0011
Rifkin R. (CIT0039) 2004; 5
Zanni L. (CIT0049) 2006; 7
CIT0014
CIT0013
CIT0016
CIT0015
CIT0018
CIT0019
Dietterich T. G. (CIT0017) 1995; 2
CIT0021
CIT0020
CIT0023
CIT0022
Murtagh B. A. (CIT0034) 1998
CIT0025
Platt J. C. (CIT0038) 2000; 12
CIT0024
CIT0027
Collins M. (CIT0009) 2008; 9
CIT0026
CIT0029
Vapnik V. N. (CIT0045) 1995
References_xml – volume: 9
  start-page: 1775
  year: 2008
  ident: CIT0009
  publication-title: J. Mach. Learn. Res.
– ident: CIT0023
  doi: 10.1109/TNN.2002.1000139
– volume-title: Minos 5.5. User's guide
  year: 1998
  ident: CIT0034
– ident: CIT0005
  doi: 10.1109/ICPR.1994.576879
– ident: CIT0011
– volume-title: Libsvm
  ident: CIT0008
  doi: 10.1145/1961189.1961199
– ident: CIT0029
– volume-title: Optimization Theory and Methods: Nonlinear Programming
  year: 2006
  ident: CIT0041
– ident: CIT0031
– ident: CIT0025
– ident: CIT0002
– ident: CIT0035
– ident: CIT0004
  doi: 10.1145/130385.130401
– ident: CIT0037
– ident: CIT0012
– ident: CIT0047
– volume: 2
  start-page: 263
  year: 1995
  ident: CIT0017
  publication-title: J. Artif. Intell. Res.
  doi: 10.1613/jair.105
– ident: CIT0032
– ident: CIT0022
– ident: CIT0016
  doi: 10.1007/s10107-005-0595-2
– ident: CIT0027
  doi: 10.1007/978-3-642-76153-9_5
– ident: CIT0036
  doi: 10.1007/BF01585748
– ident: CIT0019
– volume-title: The Nature of Statistical Learning Theory
  year: 1995
  ident: CIT0045
  doi: 10.1007/978-1-4757-2440-0
– ident: CIT0042
  doi: 10.1145/1281192.1281270
– ident: CIT0020
  doi: 10.1162/153244302320884605
– ident: CIT0040
  doi: 10.1017/CBO9780511809682
– ident: CIT0006
  doi: 10.1023/A:1008663629662
– ident: CIT0013
– ident: CIT0014
  doi: 10.1162/15324430260185628
– volume: 5
  start-page: 101
  year: 2004
  ident: CIT0039
  publication-title: J. Mach. Learn. Res.
– ident: CIT0048
  doi: 10.1016/S0167-8191(03)00021-8
– ident: CIT0003
  doi: 10.1137/S1052623497330963
– ident: CIT0010
  doi: 10.1162/15324430152733142
– volume: 12
  start-page: 547
  volume-title: Advances in Neural Information Processing Systems
  year: 2000
  ident: CIT0038
– start-page: 255
  volume-title: Advances in Kernel Methods – Support Vector Learning
  year: 1999
  ident: CIT0028
– ident: CIT0018
– ident: CIT0001
  doi: 10.1162/15324430152733133
– ident: CIT0024
  doi: 10.7551/mitpress/1130.003.0015
– ident: CIT0044
  doi: 10.1080/10556789908805759
– ident: CIT0033
  doi: 10.1098/rsta.1909.0016
– volume: 7
  start-page: 1467
  year: 2006
  ident: CIT0049
  publication-title: J. Mach. Learn. Res.
– ident: CIT0021
  doi: 10.1145/641876.641880
– volume: 6
  start-page: 1453
  year: 2005
  ident: CIT0043
  publication-title: J. Mach. Learn. Res.
– volume-title: Statistical Learning Theory
  year: 1998
  ident: CIT0046
– ident: CIT0030
– ident: CIT0026
  doi: 10.1145/1401890.1401942
– ident: CIT0015
  doi: 10.1023/A:1013637720281
– ident: CIT0007
  doi: 10.1007/0-387-30065-1_4
SSID ssj0004146
Score 1.8994557
Snippet We present a decomposition method for training Crammer and Singer's multiclass kernel-based vector machine model. A new working set selection rule is proposed....
SourceID proquest
crossref
informaworld
SourceType Aggregation Database
Enrichment Source
Index Database
Publisher
StartPage 431
SubjectTerms Algorithms
decomposition method
Kernel-based vector machines
parallel algorithm
projected gradient
Studies
Title A parallel decomposition algorithm for training multiclass kernel-based vector machines
URI https://www.tandfonline.com/doi/abs/10.1080/10556788.2011.556633
https://www.proquest.com/docview/883238511
Volume 26
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1Lj9MwELage4ED4imWBeQDt8hLGtuJe6x4qEJiuRRt4RLZiV3Qtl20zYLEr2dsj9uUrnjsJYrcJG48X-ZhzXxDyIvcFFzqSrJSyIqBvS6Z0VKw3FlTSaNGTSB7fn9STj6KdzM561Vc--qSzhw3P6-sK7mOVGEM5OqrZP9DspuHwgCcg3zhCBKG4z_JeJx55m6I_BdZa31yOGZgZXoxP4eo_8syZhFiG4iYPdh4fzk7sxcru2DeiLXZ97B1ny1DYiUmFaLD-gFUyhJrNbHd9DrAZQ36-4feps6efL3EGH_uLJrD0PAr7rB-0mwGSJz3dxmGvWyogIvpXsOPns7Mpa9fiN35ji2OFb55XeRzSYo2lsYjoHhPawo0BNEAi8gqvafbYzJkaOgJk0X2VTgvI5HGb6zZ8Yeb5KCA8KEYkIPx5PXn023FLNadpb-eiipV_vKqCXaclh1K2z0THvyS6V1yBwMKOo7ouEdu2NV9crtHM_mAnI5pwgndwQnd4ITCXDThhG5xQvs4oREnNOHkIZm-fTN9NWHYUIM1nIuODWXrjFGVNdJ5qjduG20q8FGUNXnhXDFqW4gPnI8LnHBCwBKY0mmr2qHlij8ig9X5yj4mtG2FceVIl1oZwbnzUTOYg1JXQrpK2kPC04LVDZLN-5dY1EPkpE3LXPtlruMyHxK2uetbJFv5y_WqL4u6Cyh1EaA1__OtR0luNX7R61qBeeM-BHly_QcfkVvb7-cpGXQXl_YZ-K2deY4g_AXw0ZO7
linkProvider Library Specific Holdings
linkToHtml http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV05T8MwFLagDMDAjSjl8MDqktZHkrFCVAXaTkWwWXFiA2qaojZl4NdjOw60IECCLUOelWc77_Ln7wFw5okmppFPESPUR9pfMyQiSpCnpPCpCMLYkj33-qxzS67vaYkmnDpYpcmhVUEUYW21-blNMbqExJ3bpo46dysYOPUzw3gZrNCQ-aaJAfb6H1cj3QUjLYGMSHl77ptRFrzTAnfpF1ttHVB7E4jy0wvcybA-y0U9fv3E6vgv3bbAhgtPYavYT9tgSWY7YH2OtHAX3LWg4QtPU5nCRBpIusN9wSh9GE-e8scR1ArBsvkEtJjF2ETpcCgnmUyRcZ0JfLEHBnBk4ZxyugcG7cvBRQe59gwoxpjkqEETJUTgS0GVIQ7DMo6Erz1eIIXXVKoZJomONpWJMhVRhGiNBFORDJKG1InxPqhk40weAJgkRCgWRiwKBMFYmRxMGxcW-YQqn8oqwOWq8NhRlxslUt5wDKflrHEza7yYtSpA71LPBXXHL-8H8wvOc1syUUV_E45_Fq2Vm4M7GzDlgTaW2AS0h38f-BSsdga9Lu9e9W9qYK0oaJsS0BGo5JOZPNYRUS5O7J5_Azfn_Tc
linkToPdf http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV07T8MwELagSAgG3ojy9MBqaOtH0rECqvKqGIpgs-zYBtRQUEkZ-PX4nAQKCJBgy5CzcrZzD_u77xDarekG5SriRDAeEe-vBdGKM1JzVkdcx80kkD2fd0Xnkp1c8-uxKn6AVUIO7XKiiGCr4ed-NK5ExO2Hno4-dcsJOP2zoHQSTQngDocijlr3vTKyqC_yEgREyuK5b0b54Jw-UJd-MdXB_7TnkSq_PIed9PdGmd5LXj6ROv5HtQU0VwSnuJXvpkU0YQdLaHaMsnAZXbUwsIWnqU2xsQBIL1BfWKU3D8O77PYee31w2XoCB8RiAjE67tvhwKYEHKfBz-G6AN8HMKd9WkG99lHvoEOK5gwkoZRlpM6N0zqOrOYOaMOoTZSOvL-Lra41nGs0jfGxpoMY0zHHmNdIC6dsbOrWp8WrqDJ4GNg1hI1h2ommEirWjFIHGZg3LUJFjLuI2yqi5aLIpCAuByVSWS_4TctZkzBrMp-1KiJvUo85cccv78fj6y2zcGDi8u4mkv4sulHuDVlYgCcZe1NJIZxd__vAO2j64rAtz467pxtoJj_NhvOfTVTJhiO75cOhTG-HHf8KFEv72w
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=A+parallel+decomposition+algorithm+for+training+multiclass+kernel-based+vector+machines&rft.jtitle=Optimization+methods+%26+software&rft.au=Niu%2C+Lingfeng&rft.au=Yuan%2C+Ya-Xiang&rft.date=2011-06-01&rft.pub=Taylor+%26+Francis&rft.issn=1055-6788&rft.eissn=1029-4937&rft.volume=26&rft.issue=3&rft.spage=431&rft.epage=454&rft_id=info:doi/10.1080%2F10556788.2011.556633&rft.externalDocID=556633
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1055-6788&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1055-6788&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1055-6788&client=summon