Classification and Regression Using an Outer Approximation Projection-Gradient Method

This paper deals with sparse feature selection and grouping for classification and regression. The classification or regression problems under consideration consists of minimizing a convex empirical risk function subject to an ℓ 1 constraint, a pairwise ℓ ∞ constraint, or a pairwise ℓ 1 constraint....

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on signal processing Ročník 65; číslo 17; s. 4635 - 4644
Hlavní autoři: Barlaud, Michel, Belhajali, Wafa, Combettes, Patrick L., Fillatre, Lionel
Médium: Journal Article
Jazyk:angličtina
Vydáno: IEEE 01.09.2017
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 This paper deals with sparse feature selection and grouping for classification and regression. The classification or regression problems under consideration consists of minimizing a convex empirical risk function subject to an ℓ 1 constraint, a pairwise ℓ ∞ constraint, or a pairwise ℓ 1 constraint. Existing work, such as the Lasso formulation, has focused mainly on Lagrangian penalty approximations, which often require ad hoc or computationally expensive procedures to determine the penalization parameter. We depart from this approach and address the constrained problem directly via a splitting method. The structure of the method is that of the classical gradient-projection algorithm, which alternates a gradient step on the objective and a projection step onto the lower level set modeling the constraint. The novelty of our approach is that the projection step is implemented via an outer approximation scheme in which the constraint set is approximated by a sequence of simple convex sets consisting of the intersection of two half-spaces. Convergence of the iterates generated by the algorithm is established for a general smooth convex minimization problem with inequality constraints. Experiments on both synthetic and biological data show that our method outperforms penalty methods.
AbstractList This paper deals with sparse feature selection and grouping for classification and regression. The classification or regression problems under consideration consists of minimizing a convex empirical risk function subject to an ℓ 1 constraint, a pairwise ℓ ∞ constraint, or a pairwise ℓ 1 constraint. Existing work, such as the Lasso formulation, has focused mainly on Lagrangian penalty approximations, which often require ad hoc or computationally expensive procedures to determine the penalization parameter. We depart from this approach and address the constrained problem directly via a splitting method. The structure of the method is that of the classical gradient-projection algorithm, which alternates a gradient step on the objective and a projection step onto the lower level set modeling the constraint. The novelty of our approach is that the projection step is implemented via an outer approximation scheme in which the constraint set is approximated by a sequence of simple convex sets consisting of the intersection of two half-spaces. Convergence of the iterates generated by the algorithm is established for a general smooth convex minimization problem with inequality constraints. Experiments on both synthetic and biological data show that our method outperforms penalty methods.
Author Combettes, Patrick L.
Barlaud, Michel
Fillatre, Lionel
Belhajali, Wafa
Author_xml – sequence: 1
  givenname: Michel
  surname: Barlaud
  fullname: Barlaud, Michel
  email: barlaud@i3s.unice.fr
  organization: I3S, Univ. Cote d'Azur, Sophia Antipolis, France
– sequence: 2
  givenname: Wafa
  surname: Belhajali
  fullname: Belhajali, Wafa
  email: wafa.ibnelhajali@gmail.com
  organization: I3S, Univ. Cote d'Azur, Sophia Antipolis, France
– sequence: 3
  givenname: Patrick L.
  surname: Combettes
  fullname: Combettes, Patrick L.
  email: plc@math.ncsu.edu
  organization: Dept. of Math., North Carolina State Univ., Raleigh, NC, USA
– sequence: 4
  givenname: Lionel
  surname: Fillatre
  fullname: Fillatre, Lionel
  email: fillatre@i3s.unice.fr
  organization: I3S, Univ. Cote d'Azur, Sophia Antipolis, France
BookMark eNp9UMFOAjEQbQwmAno38bI_sNh2Z1t6JETRBANRSLxt2u4US3CXtGuif28R4sGDp3nzMm9m3huQXtM2SMg1oyPGqLpdvSxHnDI54pIqLvgZ6TMFLKcgRS9hWhZ5OZavF2QQ45ZSBqBEn6ynOx2jd97qzrdNpps6e8ZNwESmdh19s0lktvjoMGST_T60n_79OLsM7RbtAeazoGuPTZc9YffW1pfk3OldxKtTHZL1_d1q-pDPF7PH6WSeWy6KLmdjZtFxUwu0aKzRANII4FBYK0tuDTUaFZQsOXRKUlcD5wq1cU6YEupiSOhxrw1tjAFdtQ_pu_BVMVodYqlSLNUhluoUS5KIPxLrux8_XdB-95_w5ij0iPh7R6oCAHjxDS67dAM
CODEN ITPRED
CitedBy_id crossref_primary_10_1080_02331934_2018_1474470
crossref_primary_10_1109_TSP_2025_3544463
crossref_primary_10_1287_ijoc_2022_0328
crossref_primary_10_1016_j_aeue_2018_02_003
crossref_primary_10_1186_s12859_021_04478_w
crossref_primary_10_1007_s10589_019_00060_6
crossref_primary_10_1109_TSP_2019_2924580
crossref_primary_10_1186_s12859_022_04900_x
crossref_primary_10_1007_s10107_019_01401_3
Cites_doi 10.1007/BF02868578
10.1145/1553374.1553431
10.1056/NEJMoa021967
10.1111/j.1467-9868.2005.00490.x
10.1111/j.2517-6161.1996.tb02080.x
10.1016/0041-5553(66)90114-5
10.1137/0149053
10.1007/978-3-642-15883-4_27
10.1093/bioinformatics/btn081
10.1016/j.crma.2008.03.014
10.1109/83.563316
10.1007/BF01585107
10.1137/S036301299732626X
10.1007/s10107-015-0946-6
10.1002/cpa.20303
10.1038/msb4100180
10.1137/060669498
10.1137/0152031
10.1109/TIP.2004.832922
10.18637/jss.v033.i01
10.1007/978-1-4419-9569-8_10
10.1137/050626090
10.1007/978-1-4613-3341-8_3
10.1111/j.1541-0420.2007.00843.x
10.1073/pnas.0437847100
10.1007/978-3-319-48311-5
ContentType Journal Article
DBID 97E
RIA
RIE
AAYXX
CITATION
DOI 10.1109/TSP.2017.2709262
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
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 4644
ExternalDocumentID 10_1109_TSP_2017_2709262
7934442
Genre orig-research
GroupedDBID -~X
.DC
0R~
29I
4.4
5GY
6IK
85S
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACIWK
ACNCT
AENEX
AGQYO
AHBIQ
AJQPL
AKJIK
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
AAYXX
CITATION
ID FETCH-LOGICAL-c263t-181cef2bd6ecebcba447b64243cc752cb0bae9451110f970fd4229eabff6b54d3
IEDL.DBID RIE
ISICitedReferencesCount 17
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000405394000016&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 Sat Nov 29 04:10:43 EST 2025
Tue Nov 18 21:27:22 EST 2025
Tue Aug 26 17:00:18 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 17
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c263t-181cef2bd6ecebcba447b64243cc752cb0bae9451110f970fd4229eabff6b54d3
PageCount 10
ParticipantIDs ieee_primary_7934442
crossref_primary_10_1109_TSP_2017_2709262
crossref_citationtrail_10_1109_TSP_2017_2709262
PublicationCentury 2000
PublicationDate 2017-Sept.1,-1
2017-9-1
PublicationDateYYYYMMDD 2017-09-01
PublicationDate_xml – month: 09
  year: 2017
  text: 2017-Sept.1,-1
  day: 01
PublicationDecade 2010
PublicationTitle IEEE transactions on signal processing
PublicationTitleAbbrev TSP
PublicationYear 2017
Publisher IEEE
Publisher_xml – name: IEEE
References ref35
ref13
ref12
ref15
ref36
ref14
belhajali (ref4) 2014
ref30
ref11
ref10
ref16
qian (ref31) 2013
ref19
schmidt (ref32) 2011
ref18
(ref1) 0
mairal (ref28) 2012
ref24
donoho (ref17) 2003; 100
ref26
ref25
hastie (ref22) 2004; 5
tibshirani (ref34) 1996; b58
friedman (ref21) 2010; 33
ref27
haugazeau (ref23) 1968
ref29
ref8
ref7
(ref2) 0
ref9
ref3
ref6
sra (ref33) 2012
ref5
figueiredo (ref20) 2016
References_xml – ident: ref29
  doi: 10.1007/BF02868578
– ident: ref24
  doi: 10.1145/1553374.1553431
– ident: ref36
  doi: 10.1056/NEJMoa021967
– ident: ref35
  doi: 10.1111/j.1467-9868.2005.00490.x
– volume: b58
  start-page: 267
  year: 1996
  ident: ref34
  article-title: Regression shrinkage and selection via the Lasso
  publication-title: J Roy Stat Soc
  doi: 10.1111/j.2517-6161.1996.tb02080.x
– start-page: 353
  year: 2012
  ident: ref28
  article-title: Complexity analysis of the Lasso regularization path
  publication-title: Proc 29th Int Conf Mach Learn
– ident: ref25
  doi: 10.1016/0041-5553(66)90114-5
– start-page: 1458
  year: 2011
  ident: ref32
  article-title: Convergence rates of inexact proximal-gradient methods for convex optimization
  publication-title: Proc Adv Neural Inf Process Syst
– year: 2013
  ident: ref31
  article-title: Glmnet for MatLab
– volume: 5
  start-page: 1391
  year: 2004
  ident: ref22
  article-title: The entire regularization path for the support vector machine
  publication-title: J Mach Learn Res
– start-page: 930
  year: 2016
  ident: ref20
  article-title: Ordered weighted $\ell _1$ regularized regression with strongly correlated covariates: Theoretical aspects
  publication-title: Proc 19th Int Conf Artif Intell
– ident: ref19
  doi: 10.1137/0149053
– year: 0
  ident: ref2
– year: 1968
  ident: ref23
  article-title: Sur les inéquations variationnelles et la minimisation de fonctionnelles convexes
– ident: ref30
  doi: 10.1007/978-3-642-15883-4_27
– year: 0
  ident: ref1
– ident: ref27
  doi: 10.1093/bioinformatics/btn081
– ident: ref6
  doi: 10.1016/j.crma.2008.03.014
– ident: ref9
  doi: 10.1109/83.563316
– start-page: 232
  year: 2014
  ident: ref4
  article-title: Boosting stochastic Newton with entropy constraint for large-scale image classification
  publication-title: Proc 22nd Int Conf Pattern Recognit
– ident: ref7
  doi: 10.1007/BF01585107
– ident: ref10
  doi: 10.1137/S036301299732626X
– ident: ref15
  doi: 10.1007/s10107-015-0946-6
– ident: ref16
  doi: 10.1002/cpa.20303
– ident: ref8
  doi: 10.1038/msb4100180
– year: 2012
  ident: ref33
  publication-title: Optimization for Machine Learning
– ident: ref12
  doi: 10.1137/060669498
– ident: ref18
  doi: 10.1137/0152031
– ident: ref11
  doi: 10.1109/TIP.2004.832922
– volume: 33
  start-page: 1
  year: 2010
  ident: ref21
  article-title: Regularization paths for generalized linear models via coordinate descent
  publication-title: J Stat Softw
  doi: 10.18637/jss.v033.i01
– ident: ref13
  doi: 10.1007/978-1-4419-9569-8_10
– ident: ref14
  doi: 10.1137/050626090
– ident: ref26
  doi: 10.1007/978-1-4613-3341-8_3
– ident: ref5
  doi: 10.1111/j.1541-0420.2007.00843.x
– volume: 100
  start-page: 2197
  year: 2003
  ident: ref17
  article-title: Optimally sparse representation in general (nonorthogonal) dictionaries via $\ell ^1$ minimization
  publication-title: Proc Nat Acad Sci USA
  doi: 10.1073/pnas.0437847100
– ident: ref3
  doi: 10.1007/978-3-319-48311-5
SSID ssj0014496
Score 2.3617165
Snippet This paper deals with sparse feature selection and grouping for classification and regression. The classification or regression problems under consideration...
SourceID crossref
ieee
SourceType Enrichment Source
Index Database
Publisher
StartPage 4635
SubjectTerms Biological system modeling
Convergence
Convex functions
Convex optimization
Level set
Optimization
outer approximation
projection-gradient algorithm
Signal processing algorithms
Title Classification and Regression Using an Outer Approximation Projection-Gradient Method
URI https://ieeexplore.ieee.org/document/7934442
Volume 65
WOSCitedRecordID wos000405394000016&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/eLvHCXMwlV1bS8MwFA5z-KAP3qY4b-TBF8FuaZpL-zjEy4tz6AZ7K01yKgPpZHbizzdJuzJBBF9KCAmULyHnnOQ750PoMhOZ4HloY5Mstx-e8EABqCA0lHBquIhU7MUm5HAYT6fJqIWum1wYAPDkM-i5pn_LN3O9dFdlfbuXGGP2wN2QUlS5Ws2LAWNei8u6C1HAYzldPUmSpD9-GTkOl-xRSVx5vB8maE1TxZuUu93__cwe2qldRzyo1noftaA4QNtrBQU7aOI1Lh37xwOOs8LgZ3ituK4F9vwA24mfnJADHrh64l-zKnkRj6o7GdsM7heeCFbiR68vfYgmd7fjm4egFk4INBVRGVirrSGnygjQoLTKGJPKBhos0lpyqhVRGSSuMllI8kSS3DBKE8hUngvFmYmOULuYF3CMMAEqMutVEBAxM3YtickTGzWHOiaGg-mi_grLVNdVxZ24xVvqowuSpBb91KGf1uh30VUz472qqPHH2I4DvhlXY37ye_cp2nKTKwbYGWqXiyWco039Wc4-Fhd-u3wDHVe-rw
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1bS8MwFA5DBfXB2xTnNQ--CHZL06SXxyHOidscusHeSpOcykA6mZ34803SrkwQwZc2hLSUL6EnJ-c750PoKvETn6eu9k2SVF94xB0BIBxXUcKp4r4nQis2EQwG4WQSDWvopsqFAQBLPoOmadpYvprJhTkqa-m1xBjTP9x1rm-kyNaqYgaMWTUuvWHwHB4Gk2VQkkSt0cvQsLiCJg2IKZD3wwitqKpYo9LZ_d_n7KGdcvOI28Vs76MaZAdoe6WkYB2Nrcql4f9YyHGSKfwMrwXbNcOWIaA78ZORcsBtU1H8a1qkL-JhcSqjm8793FLBcty3CtOHaNy5G912nVI6wZHU93JH220JKRXKBwlCioSxQGhXg3lSBpxKQUQCkalN5pI0CkiqGKURJCJNfcGZ8o7QWjbL4BhhAtRP9L6CgB8ypWeTqDTSfrMrQ6I4qAZqLbGMZVlX3MhbvMXWvyBRrNGPDfpxiX4DXVdPvBc1Nf4YWzfAV-NKzE9-775Em91Rvxf3HgaPp2jLvKjgg52htXy-gHO0IT_z6cf8wi6dbzMMwfY
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=Classification+and+Regression+Using+an+Outer+Approximation+Projection-Gradient+Method&rft.jtitle=IEEE+transactions+on+signal+processing&rft.au=Barlaud%2C+Michel&rft.au=Belhajali%2C+Wafa&rft.au=Combettes%2C+Patrick+L.&rft.au=Fillatre%2C+Lionel&rft.date=2017-09-01&rft.pub=IEEE&rft.issn=1053-587X&rft.volume=65&rft.issue=17&rft.spage=4635&rft.epage=4644&rft_id=info:doi/10.1109%2FTSP.2017.2709262&rft.externalDocID=7934442
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