Flexible Clustered Multi-Task Learning by Learning Representative Tasks

Multi-task learning (MTL) methods have shown promising performance by learning multiple relevant tasks simultaneously, which exploits to share useful information across relevant tasks. Among various MTL methods, clustered multi-task learning (CMTL) assumes that all tasks can be clustered into groups...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE transactions on pattern analysis and machine intelligence Ročník 38; číslo 2; s. 266 - 278
Hlavní autori: Zhou, Qiang, Zhao, Qi
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States IEEE 01.02.2016
Predmet:
ISSN:0162-8828, 1939-3539, 2160-9292, 1939-3539
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Multi-task learning (MTL) methods have shown promising performance by learning multiple relevant tasks simultaneously, which exploits to share useful information across relevant tasks. Among various MTL methods, clustered multi-task learning (CMTL) assumes that all tasks can be clustered into groups and attempts to learn the underlying cluster structure from the training data. In this paper, we present a new approach for CMTL, called flexible clustered multi-task (FCMTL), in which the cluster structure is learned by identifying representative tasks. The new approach allows an arbitrary task to be described by multiple representative tasks, effectively soft-assigning a task to multiple clusters with different weights. Unlike existing counterpart, the proposed approach is more flexible in that (a) it does not require clusters to be disjoint, (b) tasks within one particular cluster do not have to share information to the same extent, and (c) the number of clusters is automatically inferred from data. Computationally, the proposed approach is formulated as a row-sparsity pursuit problem. We validate the proposed FCMTL on both synthetic and real-world data sets, and empirical results demonstrate that it outperforms many existing MTL methods.
AbstractList Multi-task learning (MTL) methods have shown promising performance by learning multiple relevant tasks simultaneously, which exploits to share useful information across relevant tasks. Among various MTL methods, clustered multi-task learning (CMTL) assumes that all tasks can be clustered into groups and attempts to learn the underlying cluster structure from the training data. In this paper, we present a new approach for CMTL, called flexible clustered multi-task (FCMTL), in which the cluster structure is learned by identifying representative tasks. The new approach allows an arbitrary task to be described by multiple representative tasks, effectively soft-assigning a task to multiple clusters with different weights. Unlike existing counterpart, the proposed approach is more flexible in that (a) it does not require clusters to be disjoint, (b) tasks within one particular cluster do not have to share information to the same extent, and (c) the number of clusters is automatically inferred from data. Computationally, the proposed approach is formulated as a row-sparsity pursuit problem. We validate the proposed FCMTL on both synthetic and real-world data sets, and empirical results demonstrate that it outperforms many existing MTL methods.Multi-task learning (MTL) methods have shown promising performance by learning multiple relevant tasks simultaneously, which exploits to share useful information across relevant tasks. Among various MTL methods, clustered multi-task learning (CMTL) assumes that all tasks can be clustered into groups and attempts to learn the underlying cluster structure from the training data. In this paper, we present a new approach for CMTL, called flexible clustered multi-task (FCMTL), in which the cluster structure is learned by identifying representative tasks. The new approach allows an arbitrary task to be described by multiple representative tasks, effectively soft-assigning a task to multiple clusters with different weights. Unlike existing counterpart, the proposed approach is more flexible in that (a) it does not require clusters to be disjoint, (b) tasks within one particular cluster do not have to share information to the same extent, and (c) the number of clusters is automatically inferred from data. Computationally, the proposed approach is formulated as a row-sparsity pursuit problem. We validate the proposed FCMTL on both synthetic and real-world data sets, and empirical results demonstrate that it outperforms many existing MTL methods.
Multi-task learning (MTL) methods have shown promising performance by learning multiple relevant tasks simultaneously, which exploits to share useful information across relevant tasks. Among various MTL methods, clustered multi-task learning (CMTL) assumes that all tasks can be clustered into groups and attempts to learn the underlying cluster structure from the training data. In this paper, we present a new approach for CMTL, called flexible clustered multi-task (FCMTL), in which the cluster structure is learned by identifying representative tasks. The new approach allows an arbitrary task to be described by multiple representative tasks, effectively soft-assigning a task to multiple clusters with different weights. Unlike existing counterpart, the proposed approach is more flexible in that (a) it does not require clusters to be disjoint, (b) tasks within one particular cluster do not have to share information to the same extent, and (c) the number of clusters is automatically inferred from data. Computationally, the proposed approach is formulated as a row-sparsity pursuit problem. We validate the proposed FCMTL on both synthetic and real-world data sets, and empirical results demonstrate that it outperforms many existing MTL methods.
Author Qiang Zhou
Qi Zhao
Author_xml – sequence: 1
  givenname: Qiang
  surname: Zhou
  fullname: Zhou, Qiang
– sequence: 2
  givenname: Qi
  surname: Zhao
  fullname: Zhao, Qi
BackLink https://www.ncbi.nlm.nih.gov/pubmed/26761733$$D View this record in MEDLINE/PubMed
BookMark eNp9kE1Lw0AURQep2A_9AwqSpZvUeTNJZrIsxdZCiyJ1PUySFxlNkzqTiP33JrZVcOHqvcU5F-4dkl5ZlUjIJdAxAI1v14-T1WLMKIRjFoQsBjghAwYR9WMWsx4ZUIiYLyWTfTJ07pVSCELKz0ifRSICwfmAzGcFfpqkQG9aNK5Gi5m3aora-Gvt3rwlalua8sVLdr__E24tOixrXZsP9DrQnZPTXBcOLw53RJ5nd-vpvb98mC-mk6WfcsZrP5OQ0iTXDGLIAikgzzTQiAcaEsHTUKNIYip5AoGMUxohS7IUNWYsDJiOOB-Rm33u1lbvDbpabYxLsSh0iVXjFIiItrGCiRa9PqBNssFMba3ZaLtTx_ItIPdAaivnLOYqNV2nqqytNoUCqrqd1ffOqttZHXZuVfZHPab_K13tJYOIP4KAkAYQ8i-Cbogk
CODEN ITPIDJ
CitedBy_id crossref_primary_10_1109_TNNLS_2017_2651068
crossref_primary_10_1007_s12021_018_9381_1
crossref_primary_10_1016_j_compag_2018_10_019
crossref_primary_10_1016_j_patcog_2018_10_023
crossref_primary_10_1016_j_neunet_2019_02_007
crossref_primary_10_1016_j_compchemeng_2021_107320
crossref_primary_10_1109_TMM_2021_3055959
crossref_primary_10_1016_j_ifacol_2020_12_020
crossref_primary_10_1109_TPAMI_2017_2688363
crossref_primary_10_1016_j_neucom_2021_12_048
crossref_primary_10_3390_photonics12040324
crossref_primary_10_1016_j_neucom_2024_127259
crossref_primary_10_3390_s17102218
crossref_primary_10_1109_TMM_2022_3147664
crossref_primary_10_1016_j_csda_2024_107956
crossref_primary_10_1109_TKDE_2024_3372462
crossref_primary_10_1007_s00530_017_0534_0
crossref_primary_10_1016_j_neucom_2024_129136
crossref_primary_10_1016_j_neunet_2024_106619
crossref_primary_10_1016_j_neucom_2023_02_023
crossref_primary_10_1109_TKDE_2021_3070203
crossref_primary_10_1109_TCYB_2018_2864107
crossref_primary_10_1109_TPAMI_2021_3058852
crossref_primary_10_1109_TNNLS_2020_3028453
crossref_primary_10_1109_TFUZZ_2021_3062691
crossref_primary_10_1109_TKDE_2019_2937026
crossref_primary_10_1016_j_patcog_2018_12_018
crossref_primary_10_1016_j_neucom_2023_126237
crossref_primary_10_1109_TNNLS_2020_3026532
crossref_primary_10_1007_s00521_022_07126_3
crossref_primary_10_1109_JLT_2022_3224797
crossref_primary_10_1109_TPAMI_2020_2991344
crossref_primary_10_1007_s10489_022_04020_2
crossref_primary_10_1109_ACCESS_2024_3376441
crossref_primary_10_1109_TNNLS_2020_3042500
crossref_primary_10_1109_TSP_2021_3078625
crossref_primary_10_1007_s11222_024_10550_1
crossref_primary_10_1109_JPHOT_2021_3056471
Cites_doi 10.1109/TPAMI.2008.297
10.1145/1553374.1553431
10.1145/2339530.2339672
10.1023/A:1007379606734
10.1145/1014052.1014067
10.1007/s10994-007-5040-8
10.1109/TNN.2010.2095882
10.1109/TKDE.2009.142
10.1109/TPAMI.2012.189
10.1109/TNNLS.2012.2200262
10.1109/CVPR.2010.5540018
10.1016/j.neucom.2013.02.024
10.1109/CVPR.2012.6247852
10.1007/s10107-012-0530-2
10.1093/bioinformatics/btm611
10.1137/080716542
10.1145/2020408.2020423
10.1109/TNN.2011.2157521
10.1111/j.1467-9868.2005.00532.x
10.1007/s10107-012-0629-5
10.1145/2020408.2020549
10.1561/2200000016
10.1109/CVPR.2010.5539975
10.1162/089976603762553013
10.1017/CBO9780511804441
10.1109/TPAMI.2007.1055
10.1145/1835804.1835952
10.1145/1102351.1102479
10.1109/CVPR.2005.177
ContentType Journal Article
DBID 97E
RIA
RIE
AAYXX
CITATION
NPM
7X8
DOI 10.1109/TPAMI.2015.2452911
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
PubMed
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic

PubMed
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 Electronic Library (IEL)
  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 Engineering
Computer Science
EISSN 2160-9292
1939-3539
EndPage 278
ExternalDocumentID 26761733
10_1109_TPAMI_2015_2452911
7150415
Genre orig-research
Research Support, Non-U.S. Gov't
Journal Article
GrantInformation_xml – fundername: Defense Innovative Research Programme
  grantid: 9014100596
GroupedDBID ---
-DZ
-~X
.DC
0R~
29I
4.4
53G
5GY
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACGFS
ACIWK
ACNCT
AENEX
AGQYO
AGSQL
AHBIQ
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ASUFR
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
E.L
EBS
EJD
F5P
HZ~
IEDLZ
IFIPE
IPLJI
JAVBF
LAI
M43
MS~
O9-
OCL
P2P
PQQKQ
RIA
RIE
RNS
RXW
TAE
TN5
UHB
~02
AAYXX
CITATION
5VS
9M8
ABFSI
ADRHT
AETEA
AETIX
AI.
AIBXA
AKJIK
ALLEH
FA8
H~9
IBMZZ
ICLAB
IFJZH
NPM
RIG
RNI
RZB
VH1
XJT
7X8
ID FETCH-LOGICAL-c323t-d81c0bfa2191d4871fda10634a1b73c5ae7b9083b1489c06e2bdceaed2542a633
IEDL.DBID RIE
ISICitedReferencesCount 61
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000369989600006&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0162-8828
1939-3539
IngestDate Thu Oct 02 12:17:12 EDT 2025
Mon Jul 21 06:07:33 EDT 2025
Sat Nov 29 05:15:56 EST 2025
Tue Nov 18 22:07:02 EST 2025
Wed Aug 27 02:47:51 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 2
Keywords representative task
group sparsity
Clustered multi-task learning
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c323t-d81c0bfa2191d4871fda10634a1b73c5ae7b9083b1489c06e2bdceaed2542a633
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
PMID 26761733
PQID 1760871727
PQPubID 23479
PageCount 13
ParticipantIDs pubmed_primary_26761733
proquest_miscellaneous_1760871727
ieee_primary_7150415
crossref_citationtrail_10_1109_TPAMI_2015_2452911
crossref_primary_10_1109_TPAMI_2015_2452911
PublicationCentury 2000
PublicationDate 2016-Feb.-1
2016-2-1
2016-Feb
20160201
PublicationDateYYYYMMDD 2016-02-01
PublicationDate_xml – month: 02
  year: 2016
  text: 2016-Feb.-1
  day: 01
PublicationDecade 2010
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle IEEE transactions on pattern analysis and machine intelligence
PublicationTitleAbbrev TPAMI
PublicationTitleAlternate IEEE Trans Pattern Anal Mach Intell
PublicationYear 2016
Publisher IEEE
Publisher_xml – name: IEEE
References zhang (ref58) 0
ref13
ref56
ref12
ref59
ref15
elhamifar (ref18) 0
ref53
ref11
ref54
ref10
gong (ref23) 0
ref17
kim (ref33) 0
ref16
bonilla (ref6) 0
ref19
passos (ref38) 0
romera-paredes (ref43) 0
ref47
zhong (ref60) 0
ref41
welinder (ref50) 2010
quadrianto (ref40) 0
jacob (ref26) 0
ref49
zhang (ref55) 0
lee (ref35) 0
ref8
jalali (ref29) 0
ref9
gong (ref25) 2013; 14
ref3
chen (ref14) 0
ref5
liu (ref36) 0
romera-paredes (ref42) 0
kang (ref30) 0
ref37
ref31
ref32
wah (ref48) 2011
xue (ref51) 2007; 8
ref39
yang (ref52) 0
schwaighofer (ref44) 2004
bakker (ref4) 2003; 4
zhou (ref61) 0
zhang (ref57) 2013
ref24
evgeniou (ref20) 2005; 6
ref22
ref21
ref28
ref27
bonilla (ref7) 0
kumar (ref34) 0
solnon (ref45) 2012; 13
ref62
ando (ref1) 2005; 6
argyriou (ref2) 0
sra (ref46) 2012
References_xml – ident: ref39
  doi: 10.1109/TPAMI.2008.297
– ident: ref27
  doi: 10.1145/1553374.1553431
– start-page: 733
  year: 0
  ident: ref55
  article-title: A convex formulation for learning task relationships in multi-task learning
  publication-title: Proc 26th Conf Uncertainty Artif Intell
– ident: ref24
  doi: 10.1145/2339530.2339672
– year: 2010
  ident: ref50
  article-title: Caltech-UCSD Birds 200
– ident: ref10
  doi: 10.1023/A:1007379606734
– ident: ref21
  doi: 10.1145/1014052.1014067
– start-page: 521
  year: 0
  ident: ref30
  article-title: Learning with whom to share in multi-task feature learning
  publication-title: Proc 28th Int Conf Mach Learn
– ident: ref3
  doi: 10.1007/s10994-007-5040-8
– start-page: 702
  year: 0
  ident: ref61
  article-title: Clustered multi-task learning via alternating structure optimization
  publication-title: Proc Adv Neural Inf Process Syst
– year: 0
  ident: ref7
  article-title: Multi-task Gaussian process prediction
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref17
  doi: 10.1109/TNN.2010.2095882
– ident: ref31
  doi: 10.1109/TKDE.2009.142
– ident: ref12
  doi: 10.1109/TPAMI.2012.189
– ident: ref59
  doi: 10.1109/TNNLS.2012.2200262
– ident: ref49
  doi: 10.1109/CVPR.2010.5540018
– start-page: 1383
  year: 0
  ident: ref34
  article-title: Learning task grouping and overlap in multi-task learning
  publication-title: Proc 29th Int Conf Mach Learn
– volume: 8
  start-page: 35
  year: 2007
  ident: ref51
  article-title: Multi-task learning for classification with Dirichlet process priors
  publication-title: J Mach Learn Res
– ident: ref16
  doi: 10.1016/j.neucom.2013.02.024
– start-page: 105
  year: 0
  ident: ref14
  article-title: Smoothing proximal gradient method for general structured sparse learning
  publication-title: Proc Conf Uncertainty Artif Intell
– year: 0
  ident: ref2
  article-title: Multi-task feature learning
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref19
  doi: 10.1109/CVPR.2012.6247852
– ident: ref22
  doi: 10.1007/s10107-012-0530-2
– ident: ref28
  doi: 10.1093/bioinformatics/btm611
– start-page: 1103
  year: 0
  ident: ref38
  article-title: Flexible modeling of latent task structures in multitask learning
  publication-title: Proc Int Conf Mach Learn
– year: 2004
  ident: ref44
  article-title: Learning Gaussian process kernels via hierarchical Bayes
  publication-title: Proc Adv Neural Inf Process Syst
– start-page: 339
  year: 0
  ident: ref36
  article-title: Multi-task feature learning via efficient l2, 1-norm minimization
  publication-title: Proc Conf Uncertainty Artif Intell
– start-page: 19
  year: 0
  ident: ref18
  article-title: Finding exemplars from pairwise dissimilarities via simultaneous sparse recovery
  publication-title: Proc Adv Neural Inf Process Syst
– volume: 14
  start-page: 2979
  year: 2013
  ident: ref25
  article-title: Multi-stage multi-task feature learning
  publication-title: J Mach Learn Res
– volume: 6
  start-page: 615
  year: 2005
  ident: ref20
  article-title: Learning multiple tasks with kernel methods
  publication-title: J Mach Learn Res
– year: 0
  ident: ref35
  article-title: Adaptive multi-task lasso: With application to eqtl detection
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref5
  doi: 10.1137/080716542
– ident: ref13
  doi: 10.1145/2020408.2020423
– ident: ref41
  doi: 10.1109/TNN.2011.2157521
– volume: 6
  start-page: 1817
  year: 2005
  ident: ref1
  article-title: A framework for learning predictive structures from multiple tasks and unlabeled data
  publication-title: J Mach Learn Res
– year: 0
  ident: ref23
  article-title: Multi-stage multi-task feature learning
  publication-title: Proc Adv Neural Inf Process Syst
– year: 2012
  ident: ref46
  publication-title: Optimization for Machine Learning
– start-page: 1444
  year: 0
  ident: ref43
  article-title: Multilinear multitask learning
  publication-title: Proc 30th Int Conf Mach Learn
– ident: ref54
  doi: 10.1111/j.1467-9868.2005.00532.x
– start-page: 543
  year: 0
  ident: ref33
  article-title: Tree-guided group lasso for multi-task regression with structured sparsity
  publication-title: Proc 27th Int Conf Mach Learn
– ident: ref37
  doi: 10.1007/s10107-012-0629-5
– ident: ref62
  doi: 10.1145/2020408.2020549
– ident: ref8
  doi: 10.1561/2200000016
– year: 0
  ident: ref42
  article-title: Exploiting unrelated tasks in multi-task learning
  publication-title: Proc Int Conf Artif Intell Statist
– year: 0
  ident: ref58
  article-title: Probabilistic multi-task feature selection
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref56
  doi: 10.1109/CVPR.2010.5539975
– ident: ref32
  doi: 10.1162/089976603762553013
– start-page: 2151
  year: 0
  ident: ref52
  article-title: Heterogeneous multitask learning with joint sparsity constraints
  publication-title: Proc Adv Neural Inf Process Syst
– start-page: 43
  year: 0
  ident: ref6
  article-title: Kernel multi-task learning using task-specific features
  publication-title: Proc 11th Int Conf Artif Intell Statist
– ident: ref9
  doi: 10.1017/CBO9780511804441
– start-page: 49
  year: 0
  ident: ref60
  article-title: Convex multitask learning with flexible task clusters
  publication-title: Proc 29th Int Conf Mach Learn
– year: 0
  ident: ref29
  article-title: A dirty model for multi-task learning
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref47
  doi: 10.1109/TPAMI.2007.1055
– start-page: 1917
  year: 2013
  ident: ref57
  article-title: Learning high-order task relationships in multi-task learning
  publication-title: Proc 23rd Int Joint Conf Artif Intell
– ident: ref11
  doi: 10.1145/1835804.1835952
– volume: 4
  start-page: 83
  year: 2003
  ident: ref4
  article-title: Task clustering and gating for Bayesian multitask learning
  publication-title: J Mach Learn Res
– ident: ref53
  doi: 10.1145/1102351.1102479
– volume: 13
  start-page: 2773
  year: 2012
  ident: ref45
  article-title: Multi-task regression using minimal penalties
  publication-title: J Mach Learn Res
– year: 0
  ident: ref26
  article-title: Clustered multi-task learning: A convex formulation
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref15
  doi: 10.1109/CVPR.2005.177
– year: 2011
  ident: ref48
  article-title: The Caltech-UCSD Birds-200-2011 Dataset
– year: 0
  ident: ref40
  article-title: Multitask learning without label correspondences
  publication-title: Proc Adv Neural Inf Process Syst
SSID ssj0014503
Score 2.457809
Snippet Multi-task learning (MTL) methods have shown promising performance by learning multiple relevant tasks simultaneously, which exploits to share useful...
SourceID proquest
pubmed
crossref
ieee
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 266
SubjectTerms Clustered Multi-Task Learning
Covariance matrices
Group Sparsity
Kernel
Learning systems
Optimization
Representative Task
Robustness
Training data
Visualization
Title Flexible Clustered Multi-Task Learning by Learning Representative Tasks
URI https://ieeexplore.ieee.org/document/7150415
https://www.ncbi.nlm.nih.gov/pubmed/26761733
https://www.proquest.com/docview/1760871727
Volume 38
WOSCitedRecordID wos000369989600006&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: 2160-9292
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0014503
  issn: 0162-8828
  databaseCode: RIE
  dateStart: 19790101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8QwEB5UPOjBVdfH-qKCN602Tdu0R1lc9aCIrLC3ksesiMuu7Av8907ShwoqeMshSdvMJPmmM_MNwAkh2FTFKvZ5n6EfZZGlvI3oMJRaZpYvXIXKFZsQ9_dpr5c9LMBZnQuDiC74DM9t0_nyzUjP7K-yC0HoxWWULwohilyt2mMQxa4KMiEY2uFkRlQJMkF20X24vLu1UVzxufMzMlseJkzIgBecf7uPXIGV37Gmu3M6jf-97TqsldjSuyyUYQMWcLgJjapug1du401Y_UJC2ITrjuXEVAP02oOZpU1A47m0XL8rJ69eScD67Kn3z_aji58t0pbm6NmOky146lx12zd-WV7B1zzkU9-kTAeqL-nMYobsFtY3kgxEHkmmBNexRKEyQmiKLKZMBwmGymiUaMimDGXC-TYsDUdD3AWvLyKCv7GmyVQUM54qkn-aoJE6CZk2LWDVIue65B63JTAGubNBgix3MsqtjPJSRi04rce8Fcwbf_ZuWgnUPcvFb8FxJcuc9o11hsghjmaTnIkkoG8m-NaCnULI9eBKN_Z-nnQfVujRZez2ASxNxzM8hGU9n75MxkeknL30yCnnByI33OU
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1ZSwMxEB5EBfXB-6jnCr7p1s0mez2KWBW1iFTwbckxFbG0YlvBf-8ke6iggm95SEI2M0m-2Zn5BuCAEGyqIhX5vMvQF5mwlLeCLkOpZWb5wlWoXLGJpN1OHx6y2wk4qnNhENEFn2HTNp0v3wz02P4qO04IvbiM8qlIiJAV2Vq1z0BErg4yYRg642RIVCkyQXbcuT25ubRxXFHTeRqZLRATxmTCJ5x_e5FciZXf0aZ7dVoL_1vvIsyX6NI7KdRhCSawvwwLVeUGrzzIyzD3hYZwBc5blhVT9dA77Y0tcQIazyXm-h05fPZKCtZHT71_tu9cBG2RuPSGnu04XIX71lnn9MIvCyz4mod85JuU6UB1Jd1azJDlwrpGkonIhWQq4TqSmKiMMJoimynTQYyhMholGrIqQxlzvgaT_UEfN8DrJoIAcKRpMiUixlNFGpDGaKSOQ6ZNA1i1ybku2cdtEYxe7qyQIMudjHIro7yUUQMO6zEvBffGn71XrATqnuXmN2C_kmVOJ8e6Q2QfB-NhzpI4oG8mANeA9ULI9eBKNzZ_nnQPZi46N9f59WX7agtmaRllJPc2TI5ex7gD0_pt9DR83XUq-gGS299E
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=Flexible+Clustered+Multi-Task+Learning+by+Learning+Representative+Tasks&rft.jtitle=IEEE+transactions+on+pattern+analysis+and+machine+intelligence&rft.au=Zhou%2C+Qiang&rft.au=Zhao%2C+Qi&rft.date=2016-02-01&rft.issn=0162-8828&rft.eissn=2160-9292&rft.volume=38&rft.issue=2&rft.spage=266&rft.epage=278&rft_id=info:doi/10.1109%2FTPAMI.2015.2452911&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TPAMI_2015_2452911
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0162-8828&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0162-8828&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0162-8828&client=summon