Characterization of the equivalence of robustification and regularization in linear and matrix regression

•We characterize the connection between robust optimization and regularization.•We extend the characterization to new settings such as matrix completion.•Robust optimization and regularization are not always equivalent. The notion of developing statistical methods in machine learning which are robus...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:European journal of operational research Jg. 270; H. 3; S. 931 - 942
Hauptverfasser: Bertsimas, Dimitris, Copenhaver, Martin S.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.11.2018
Schlagworte:
ISSN:0377-2217, 1872-6860
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract •We characterize the connection between robust optimization and regularization.•We extend the characterization to new settings such as matrix completion.•Robust optimization and regularization are not always equivalent. The notion of developing statistical methods in machine learning which are robust to adversarial perturbations in the underlying data has been the subject of increasing interest in recent years. A common feature of this work is that the adversarial robustification often corresponds exactly to regularization methods which appear as a loss function plus a penalty. In this paper we deepen and extend the understanding of the connection between robustification and regularization (as achieved by penalization) in regression problems. Specifically,(a)In the context of linear regression, we characterize precisely under which conditions on the model of uncertainty used and on the loss function penalties robustification and regularization are equivalent.(b)We extend the characterization of robustification and regularization to matrix regression problems (matrix completion and Principal Component Analysis).
AbstractList •We characterize the connection between robust optimization and regularization.•We extend the characterization to new settings such as matrix completion.•Robust optimization and regularization are not always equivalent. The notion of developing statistical methods in machine learning which are robust to adversarial perturbations in the underlying data has been the subject of increasing interest in recent years. A common feature of this work is that the adversarial robustification often corresponds exactly to regularization methods which appear as a loss function plus a penalty. In this paper we deepen and extend the understanding of the connection between robustification and regularization (as achieved by penalization) in regression problems. Specifically,(a)In the context of linear regression, we characterize precisely under which conditions on the model of uncertainty used and on the loss function penalties robustification and regularization are equivalent.(b)We extend the characterization of robustification and regularization to matrix regression problems (matrix completion and Principal Component Analysis).
Author Copenhaver, Martin S.
Bertsimas, Dimitris
Author_xml – sequence: 1
  givenname: Dimitris
  orcidid: 0000-0002-1985-1003
  surname: Bertsimas
  fullname: Bertsimas, Dimitris
  email: dbertsim@mit.edu
  organization: Sloan School of Management and Operations Research Center, MIT, United States
– sequence: 2
  givenname: Martin S.
  orcidid: 0000-0002-9988-260X
  surname: Copenhaver
  fullname: Copenhaver, Martin S.
  email: mcopen@mit.edu
  organization: Operations Research Center, MIT, United States
BookMark eNp9kE1PwzAMhiM0JMbgD3DqH2hx0o-0Ehc08SVN4gLnKE1c5qprIckm4NfTbogDh50s2e9j2c85m_VDj4xdcUg48OK6TbAdXCKAywTSBHJ-wua8lCIuygJmbA6plLEQXJ6xc-9bAOA5z-eMlmvttAno6FsHGvpoaKKwxgg_trTTHfYGp5Yb6q0P1JA5pHRvI4dv207_gdRHHfWo3X640cHR55Rx6P04v2Cnje48Xv7WBXu9v3tZPsar54en5e0qNlkmQpxZUVVNLcvc5mUpdCUtb7AoubQ6zesUrGnQphpNXWPdAFSZ0AUKDgJEOmYWrDzsNW7w3mGjDIX9hcFp6hQHNSlTrZqUqUmZglSNykZU_EPfHW20-zoO3RwgHJ_aETrlDU3aLDk0QdmBjuE_QKyLTQ
CitedBy_id crossref_primary_10_1007_s10107_019_01419_7
crossref_primary_10_1016_j_ejor_2025_02_005
crossref_primary_10_1287_ijoo_2020_0038
crossref_primary_10_1007_s11590_021_01770_9
crossref_primary_10_1109_TSG_2022_3211546
crossref_primary_10_1007_s10994_021_06123_2
crossref_primary_10_1016_j_orl_2020_02_008
crossref_primary_10_1007_s11071_022_08178_9
crossref_primary_10_1016_j_ifacol_2025_07_191
crossref_primary_10_1016_j_sigpro_2021_108439
crossref_primary_10_1016_j_arcontrol_2021_09_005
crossref_primary_10_1002_for_2857
crossref_primary_10_1016_j_cej_2021_133032
crossref_primary_10_1111_stan_12253
crossref_primary_10_1007_s11831_022_09733_8
crossref_primary_10_1016_j_aca_2020_08_054
crossref_primary_10_1109_LCSYS_2020_3043591
crossref_primary_10_1109_TEMPR_2025_3530266
crossref_primary_10_1109_TPWRS_2024_3442954
crossref_primary_10_1007_s10994_020_05868_6
crossref_primary_10_1287_mnsc_2023_04164
crossref_primary_10_1007_s12532_018_0153_6
crossref_primary_10_1007_s10107_022_01796_6
crossref_primary_10_1287_opre_2022_2383
crossref_primary_10_1007_s00362_020_01214_z
crossref_primary_10_1287_ijoc_2022_0022
crossref_primary_10_1016_j_ejor_2023_11_036
crossref_primary_10_1111_poms_13239
crossref_primary_10_1214_20_STS701REJ
crossref_primary_10_1007_s10994_024_06577_0
crossref_primary_10_1016_j_ejor_2021_05_049
crossref_primary_10_2139_ssrn_3182422
crossref_primary_10_1007_s10994_021_06085_5
crossref_primary_10_1007_s13042_019_01044_y
crossref_primary_10_1214_20_STS809
crossref_primary_10_2139_ssrn_3623006
crossref_primary_10_1017_jpr_2019_49
crossref_primary_10_1007_s43069_022_00151_x
crossref_primary_10_1016_j_jco_2025_101979
crossref_primary_10_1109_MCS_2023_3291638
crossref_primary_10_1287_ijoc_2022_1211
crossref_primary_10_1016_j_orl_2021_01_002
crossref_primary_10_1016_j_ejor_2020_08_045
crossref_primary_10_1016_j_ifacol_2021_08_357
crossref_primary_10_1080_02664763_2020_1765323
crossref_primary_10_1287_msom_2020_0893
crossref_primary_10_1007_s10994_021_06036_0
crossref_primary_10_1016_j_asoc_2021_107385
crossref_primary_10_1287_ijoo_2018_0001
crossref_primary_10_1287_opre_2020_2041
crossref_primary_10_1287_opre_2019_1969
Cites_doi 10.1137/080734510
10.1198/004017004000000563
10.1111/j.1467-9868.2010.00764.x
10.1145/1970392.1970395
10.1137/070697835
10.1145/2184319.2184343
10.1007/s10260-006-0034-4
10.1016/j.jco.2009.01.002
10.1198/016214506000000096
10.1111/j.1467-9868.2005.00503.x
10.1214/aos/1176342503
10.1080/01621459.1984.10477105
10.1214/13-AOS1191
10.1016/j.sigpro.2007.04.004
10.1111/j.2517-6161.1996.tb02080.x
10.1137/0717073
10.1016/j.jmva.2004.08.002
10.1137/S0895479896298130
10.1007/BF02288367
10.1016/j.jspi.2003.12.020
10.1214/088342307000000087
10.1080/01621459.1974.10482962
10.1137/08073682X
10.1287/opre.2015.1374
10.1109/TIT.2010.2048503
ContentType Journal Article
Copyright 2017 Elsevier B.V.
Copyright_xml – notice: 2017 Elsevier B.V.
DBID AAYXX
CITATION
DOI 10.1016/j.ejor.2017.03.051
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Computer Science
Business
EISSN 1872-6860
EndPage 942
ExternalDocumentID 10_1016_j_ejor_2017_03_051
S0377221717302734
GroupedDBID --K
--M
-~X
.DC
.~1
0R~
1B1
1RT
1~.
1~5
4.4
457
4G.
5GY
5VS
6OB
7-5
71M
8P~
9JN
9JO
AAAKF
AABNK
AACTN
AAEDT
AAEDW
AAIAV
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AARIN
AAXUO
AAYFN
ABAOU
ABBOA
ABFNM
ABFRF
ABJNI
ABMAC
ABUCO
ABYKQ
ACAZW
ACDAQ
ACGFO
ACGFS
ACIWK
ACNCT
ACRLP
ACZNC
ADBBV
ADEZE
ADGUI
AEBSH
AEFWE
AEKER
AENEX
AFKWA
AFTJW
AGHFR
AGUBO
AGYEJ
AHHHB
AHZHX
AIALX
AIEXJ
AIGVJ
AIKHN
AITUG
AJOXV
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
AOUOD
APLSM
ARUGR
AXJTR
BKOJK
BKOMP
BLXMC
CS3
DU5
EBS
EFJIC
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
HAMUX
IHE
J1W
KOM
LY1
M41
MHUIS
MO0
MS~
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
PQQKQ
Q38
RIG
ROL
RPZ
RXW
SCC
SDF
SDG
SDP
SDS
SES
SPC
SPCBC
SSB
SSD
SSV
SSW
SSZ
T5K
TAE
TN5
U5U
XPP
ZMT
~02
~G-
1OL
29G
41~
9DU
AAAKG
AAQXK
AATTM
AAXKI
AAYWO
AAYXX
ABWVN
ABXDB
ACLOT
ACNNM
ACRPL
ACVFH
ADCNI
ADIYS
ADJOM
ADMUD
ADNMO
ADXHL
AEIPS
AEUPX
AFFNX
AFJKZ
AFPUW
AGQPQ
AI.
AIGII
AIIUN
AKBMS
AKRWK
AKYEP
ANKPU
APXCP
ASPBG
AVWKF
AZFZN
CITATION
EFKBS
FEDTE
FGOYB
HVGLF
HZ~
R2-
SEW
VH1
WUQ
~HD
ID FETCH-LOGICAL-c442t-4d299fb785d5882a97d1fe6817da35b30dcfed3aecbbebf00942a6e2102023da3
ISICitedReferencesCount 71
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000438660400013&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0377-2217
IngestDate Sat Nov 29 05:34:30 EST 2025
Tue Nov 18 22:11:22 EST 2025
Fri Feb 23 02:27:40 EST 2024
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 3
Keywords Statistical regression
Penalty methods
Robust optimization
Adversarial learning
Convex programming
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c442t-4d299fb785d5882a97d1fe6817da35b30dcfed3aecbbebf00942a6e2102023da3
ORCID 0000-0002-1985-1003
0000-0002-9988-260X
OpenAccessLink https://hdl.handle.net/1721.1/135747
PageCount 12
ParticipantIDs crossref_citationtrail_10_1016_j_ejor_2017_03_051
crossref_primary_10_1016_j_ejor_2017_03_051
elsevier_sciencedirect_doi_10_1016_j_ejor_2017_03_051
PublicationCentury 2000
PublicationDate 2018-11-01
PublicationDateYYYYMMDD 2018-11-01
PublicationDate_xml – month: 11
  year: 2018
  text: 2018-11-01
  day: 01
PublicationDecade 2010
PublicationTitle European journal of operational research
PublicationYear 2018
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Candès, Recht (bib0010) 2012; 55
Golub, Van Loan (bib0019) 1980; 17
Shaham, U., Yamada, Y., & Negahban, S. (2015). Understanding adversarial training: Increasing local stability of neural nets through robust optimization. arXiv preprint arXiv
Eckart, Young (bib0015) 1936; 1
Fan, Fan, Barut (bib0016) 2014; 42
Hastie, Tibshirani, Friedman (bib0023) 2009
Lewis (bib0031) 2002
Mosci, Rosasco, Santoro, Verri, Villa (bib0036) 2010
Morgenthaler (bib0035) 2007; 15
Caramanis, Mannor, Xu (bib0011) 2011
Boyd, Vandenberghe (bib0007) 2004
Huber (bib0026) 1973; 1
Lewis, Pang (bib0032) 2009; 48
Xu, Caramanis, Mannor (bib0045) 2010; 56
Ghaoui, Lebret (bib0018) 1997; 18
Markovsky, Huffel (bib0034) 2007; 87
Salibian-Barrera, Aelst, Willems (bib0040) 2005; 101
Fazel (bib0017) 2002
Kukush, Markovsky, Huffel (bib0030) 2005; 133
Rousseeuw, Leroy (bib0039) 1987
Carroll, Ruppert, Stefanski, Crainiceanu (bib0012) 2006
Croux, Ruiz-Gazen (bib0013) 2005; 95
Bertsimas, D., Gupta, V., & Kallus, N. (2017). Data-driven robust optimization. Mathematical Programming.
Hill (bib0024) 1977
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014b). Explaining and harnessing adversarial examples. arXiv preprint arXiv
Bertsimas, Brown, Caramanis (bib0004) 2011; 53
Ben-Tal, Ghaoui, Nemirovski (bib0002) 2009
Bradic, Fan, Wang (bib0008) 2011; 73
De Mol, De Vito, Rosasco (bib0014) 2009; 25
Rousseeuw (bib0038) 1984; 79
Hubert, Rousseeuw, Aelst (bib0028) 2008; 23
Zou, Hastie (bib0046) 2005; 67
Mallows (bib0033) 1975
.
Horn, Johnson (bib0025) 2013
Recht, Fazel, Parrilo (bib0037) 2010; 52
Candès, Li, Ma, Wright (bib0009) 2011; 58
Hampel (bib0022) 1974; 69
Ben-Tal, Hazan, Koren, Mannor (bib0003) 2015; 63
Hubert, Rousseeuw, den Branden (bib0029) 2005; 47
Goodfellow, Pouget-Abadie, Mirza, Xu, Warde-Farley, Ozair (bib0020) 2014
Tulabandhula, T., & Rudin, C. (2014). Robust optimization using machine learning for uncertainty sets. arXiv preprint arXiv
Bauschke, Combettes (bib0001) 2011
SIGKDD, Netflix (bib0042) 2007
Bousquet, Boucheron, Lugosi (bib0006) 2004
Tibshirani (bib0043) 1996; 58
Huber, Ronchetti (bib0027) 2009
Hill (10.1016/j.ejor.2017.03.051_bib0024) 1977
Candès (10.1016/j.ejor.2017.03.051_bib0009) 2011; 58
Recht (10.1016/j.ejor.2017.03.051_bib0037) 2010; 52
De Mol (10.1016/j.ejor.2017.03.051_bib0014) 2009; 25
Kukush (10.1016/j.ejor.2017.03.051_bib0030) 2005; 133
Markovsky (10.1016/j.ejor.2017.03.051_bib0034) 2007; 87
Boyd (10.1016/j.ejor.2017.03.051_bib0007) 2004
Fazel (10.1016/j.ejor.2017.03.051_bib0017) 2002
Bauschke (10.1016/j.ejor.2017.03.051_bib0001) 2011
SIGKDD (10.1016/j.ejor.2017.03.051_bib0042) 2007
10.1016/j.ejor.2017.03.051_bib0044
10.1016/j.ejor.2017.03.051_bib0021
10.1016/j.ejor.2017.03.051_bib0005
Mosci (10.1016/j.ejor.2017.03.051_bib0036) 2010
Goodfellow (10.1016/j.ejor.2017.03.051_bib0020) 2014
Carroll (10.1016/j.ejor.2017.03.051_bib0012) 2006
Morgenthaler (10.1016/j.ejor.2017.03.051_bib0035) 2007; 15
Caramanis (10.1016/j.ejor.2017.03.051_bib0011) 2011
Mallows (10.1016/j.ejor.2017.03.051_bib0033) 1975
Rousseeuw (10.1016/j.ejor.2017.03.051_bib0038) 1984; 79
Candès (10.1016/j.ejor.2017.03.051_bib0010) 2012; 55
Zou (10.1016/j.ejor.2017.03.051_bib0046) 2005; 67
Horn (10.1016/j.ejor.2017.03.051_bib0025) 2013
Ghaoui (10.1016/j.ejor.2017.03.051_bib0018) 1997; 18
Hastie (10.1016/j.ejor.2017.03.051_bib0023) 2009
Huber (10.1016/j.ejor.2017.03.051_bib0027) 2009
Lewis (10.1016/j.ejor.2017.03.051_bib0032) 2009; 48
10.1016/j.ejor.2017.03.051_bib0041
Bradic (10.1016/j.ejor.2017.03.051_bib0008) 2011; 73
Eckart (10.1016/j.ejor.2017.03.051_sbref0014) 1936; 1
Hubert (10.1016/j.ejor.2017.03.051_bib0029) 2005; 47
Salibian-Barrera (10.1016/j.ejor.2017.03.051_bib0040) 2005; 101
Fan (10.1016/j.ejor.2017.03.051_bib0016) 2014; 42
Ben-Tal (10.1016/j.ejor.2017.03.051_bib0002) 2009
Hampel (10.1016/j.ejor.2017.03.051_bib0022) 1974; 69
Ben-Tal (10.1016/j.ejor.2017.03.051_bib0003) 2015; 63
Croux (10.1016/j.ejor.2017.03.051_bib0013) 2005; 95
Rousseeuw (10.1016/j.ejor.2017.03.051_bib0039) 1987
Bousquet (10.1016/j.ejor.2017.03.051_bib0006) 2004
Golub (10.1016/j.ejor.2017.03.051_bib0019) 1980; 17
Huber (10.1016/j.ejor.2017.03.051_bib0026) 1973; 1
Hubert (10.1016/j.ejor.2017.03.051_bib0028) 2008; 23
Lewis (10.1016/j.ejor.2017.03.051_bib0031) 2002
Bertsimas (10.1016/j.ejor.2017.03.051_bib0004) 2011; 53
Tibshirani (10.1016/j.ejor.2017.03.051_bib0043) 1996; 58
Xu (10.1016/j.ejor.2017.03.051_bib0045) 2010; 56
References_xml – volume: 25
  start-page: 201
  year: 2009
  end-page: 230
  ident: bib0014
  article-title: Elastic-net regularization in learning theory
  publication-title: Journal of Complexity
– reference: Bertsimas, D., Gupta, V., & Kallus, N. (2017). Data-driven robust optimization. Mathematical Programming.
– year: 2006
  ident: bib0012
  article-title: Measurement error in nonlinear models: A modern perspective
– year: 2009
  ident: bib0023
  article-title: The elements of statistical learning: Data mining, inference, and prediction
– volume: 95
  start-page: 206
  year: 2005
  end-page: 226
  ident: bib0013
  article-title: High breakdown estimators for principal components: The projection-pursuit approach revisited
  publication-title: Journal of Multivariate Analysis
– year: 1987
  ident: bib0039
  article-title: Robust regression and outlier detection
– year: 2004
  ident: bib0006
  article-title: Advanced lectures on machine learning
– year: 2011
  ident: bib0011
  article-title: Optimization for machine learning
– volume: 47
  start-page: 64
  year: 2005
  end-page: 79
  ident: bib0029
  article-title: ROBPCA: A new approach to robust principal components analysis
  publication-title: Technometrics
– volume: 73
  start-page: 325
  year: 2011
  end-page: 349
  ident: bib0008
  article-title: Penalized composite quasi-likelihood for ultrahigh dimensional variable selection
  publication-title: Journal of the Royal Statistical Society, Series B
– year: 2002
  ident: bib0017
  publication-title: Matrix rank minimization with applications
– reference: Shaham, U., Yamada, Y., & Negahban, S. (2015). Understanding adversarial training: Increasing local stability of neural nets through robust optimization. arXiv preprint arXiv:
– volume: 67
  start-page: 301
  year: 2005
  end-page: 320
  ident: bib0046
  article-title: Regularization and variable selection via the elastic net
  publication-title: Journal of the Royal Statistical Society: Series B
– volume: 17
  start-page: 883
  year: 1980
  end-page: 893
  ident: bib0019
  article-title: An analysis of the total least squares problem
  publication-title: SIAM Journal of Numerical Analysis
– volume: 133
  start-page: 315
  year: 2005
  end-page: 358
  ident: bib0030
  article-title: Consistency of the structured total least squares estimator in a multivariate errors-in-variables model
  publication-title: Journal of Statistical Planning and Inference
– volume: 48
  start-page: 3080
  year: 2009
  end-page: 3104
  ident: bib0032
  article-title: Lipschitz behavior of the robust regularization
  publication-title: SIAM Journal on Control and Optimization
– volume: 63
  start-page: 628
  year: 2015
  end-page: 638
  ident: bib0003
  article-title: Oracle-based robust optimization via online learning
  publication-title: Operations Research
– volume: 23
  start-page: 92
  year: 2008
  end-page: 119
  ident: bib0028
  article-title: High-breakdown robust multivariate methods
  publication-title: Statistical Science
– volume: 1
  year: 1936
  ident: bib0015
  article-title: The approximation of one matrix by another of lower rank
  publication-title: Psychometrika
– volume: 58
  start-page: 11:1
  year: 2011
  end-page: 37
  ident: bib0009
  article-title: Robust Principal Component Analysis?
  publication-title: Journal of the ACM
– year: 2011
  ident: bib0001
  article-title: Convex analysis and monotone operator theory in Hilbert spaces
– volume: 42
  start-page: 324
  year: 2014
  end-page: 351
  ident: bib0016
  article-title: Adaptive robust variable selection
  publication-title: The Annals of Statistics
– volume: 53
  start-page: 464
  year: 2011
  end-page: 501
  ident: bib0004
  article-title: Theory and applications of robust optimization
  publication-title: SIAM Review
– reference: Tulabandhula, T., & Rudin, C. (2014). Robust optimization using machine learning for uncertainty sets. arXiv preprint arXiv:
– volume: 1
  start-page: 799
  year: 1973
  end-page: 821
  ident: bib0026
  article-title: Robust regression: Asymptotics, conjectures and Monte Carlo
  publication-title: The Annals of Statistics
– volume: 101
  start-page: 1198
  year: 2005
  end-page: 1211
  ident: bib0040
  article-title: PCA based on multivariate MM-estimators with fast and robust bootstrap
  publication-title: Journal of the American Statistical Association
– start-page: 418
  year: 2010
  end-page: 433
  ident: bib0036
  article-title: Solving structured sparsity regularization with proximal methods
  publication-title: Proceedings of the Joint european conference on machine learning and knowledge discovery in databases
– volume: 56
  start-page: 3561
  year: 2010
  end-page: 3574
  ident: bib0045
  article-title: Robust regression and Lasso
  publication-title: IEEE Transactions in Information Theory
– volume: 87
  start-page: 2283
  year: 2007
  end-page: 2302
  ident: bib0034
  article-title: Overview of total least-squares methods
  publication-title: Signal Processing
– volume: 15
  start-page: 271
  year: 2007
  end-page: 293
  ident: bib0035
  article-title: A survey of robust statistics
  publication-title: Statistical Methods and Applications
– year: 2007
  ident: bib0042
  article-title: Soft modelling by latent variables: The nonlinear iterative partial least squares (NIPALS) approach
  publication-title: Proceedings of the KDD Cup and Workshop
– reference: Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014b). Explaining and harnessing adversarial examples. arXiv preprint arXiv:
– reference: .
– year: 2004
  ident: bib0007
  article-title: Convex optimization
– volume: 69
  start-page: 383
  year: 1974
  end-page: 393
  ident: bib0022
  article-title: The influence curve and its role in robust estimation
  publication-title: Journal of the American Statistical Association
– volume: 52
  start-page: 471
  year: 2010
  end-page: 501
  ident: bib0037
  article-title: Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization
  publication-title: SIAM Review
– volume: 79
  start-page: 871
  year: 1984
  end-page: 880
  ident: bib0038
  article-title: Least median of squares regression
  publication-title: Journal of the American Statistical Association
– volume: 55
  start-page: 111
  year: 2012
  end-page: 119
  ident: bib0010
  article-title: Exact matrix completion via convex optimization
  publication-title: Communications of the ACM
– year: 1977
  ident: bib0024
  publication-title: Robust regression when there are outlires in the carriers
– start-page: 2672
  year: 2014
  end-page: 2680
  ident: bib0020
  article-title: Generative adversarial nets
  publication-title: Advances in neural information processing systems 27
– year: 2013
  ident: bib0025
  article-title: Matrix analysis
– year: 2009
  ident: bib0002
  article-title: Robust optimization
– year: 1975
  ident: bib0033
  article-title: On some topics in robustness
  publication-title: Technical Report
– volume: 18
  start-page: 1035
  year: 1997
  end-page: 1064
  ident: bib0018
  article-title: Robust solutions to least-squares problems with uncertain data
  publication-title: SIAM Journal of Matrix Analysis and Applications
– year: 2009
  ident: bib0027
  article-title: Robust statistics
– volume: 58
  start-page: 267
  year: 1996
  end-page: 288
  ident: bib0043
  article-title: Regression shrinkage and selection via the Lasso
  publication-title: Journal of the Royal Statistical Society, Series B
– year: 2002
  ident: bib0031
  article-title: Robust regularization
  publication-title: Technical Report
– volume: 53
  start-page: 464
  issue: 3
  year: 2011
  ident: 10.1016/j.ejor.2017.03.051_bib0004
  article-title: Theory and applications of robust optimization
  publication-title: SIAM Review
  doi: 10.1137/080734510
– start-page: 2672
  year: 2014
  ident: 10.1016/j.ejor.2017.03.051_bib0020
  article-title: Generative adversarial nets
– volume: 47
  start-page: 64
  year: 2005
  ident: 10.1016/j.ejor.2017.03.051_bib0029
  article-title: ROBPCA: A new approach to robust principal components analysis
  publication-title: Technometrics
  doi: 10.1198/004017004000000563
– year: 2009
  ident: 10.1016/j.ejor.2017.03.051_bib0023
– year: 2011
  ident: 10.1016/j.ejor.2017.03.051_bib0011
– year: 2013
  ident: 10.1016/j.ejor.2017.03.051_bib0025
– year: 1987
  ident: 10.1016/j.ejor.2017.03.051_bib0039
– volume: 73
  start-page: 325
  year: 2011
  ident: 10.1016/j.ejor.2017.03.051_bib0008
  article-title: Penalized composite quasi-likelihood for ultrahigh dimensional variable selection
  publication-title: Journal of the Royal Statistical Society, Series B
  doi: 10.1111/j.1467-9868.2010.00764.x
– volume: 58
  start-page: 11:1
  issue: 3
  year: 2011
  ident: 10.1016/j.ejor.2017.03.051_bib0009
  article-title: Robust Principal Component Analysis?
  publication-title: Journal of the ACM
  doi: 10.1145/1970392.1970395
– ident: 10.1016/j.ejor.2017.03.051_bib0021
– volume: 52
  start-page: 471
  issue: 3
  year: 2010
  ident: 10.1016/j.ejor.2017.03.051_bib0037
  article-title: Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization
  publication-title: SIAM Review
  doi: 10.1137/070697835
– ident: 10.1016/j.ejor.2017.03.051_bib0044
– volume: 55
  start-page: 111
  issue: 6
  year: 2012
  ident: 10.1016/j.ejor.2017.03.051_bib0010
  article-title: Exact matrix completion via convex optimization
  publication-title: Communications of the ACM
  doi: 10.1145/2184319.2184343
– year: 2002
  ident: 10.1016/j.ejor.2017.03.051_bib0017
– volume: 15
  start-page: 271
  year: 2007
  ident: 10.1016/j.ejor.2017.03.051_bib0035
  article-title: A survey of robust statistics
  publication-title: Statistical Methods and Applications
  doi: 10.1007/s10260-006-0034-4
– year: 2002
  ident: 10.1016/j.ejor.2017.03.051_bib0031
  article-title: Robust regularization
– year: 2006
  ident: 10.1016/j.ejor.2017.03.051_bib0012
– volume: 25
  start-page: 201
  issue: 2
  year: 2009
  ident: 10.1016/j.ejor.2017.03.051_bib0014
  article-title: Elastic-net regularization in learning theory
  publication-title: Journal of Complexity
  doi: 10.1016/j.jco.2009.01.002
– year: 2004
  ident: 10.1016/j.ejor.2017.03.051_bib0006
– volume: 101
  start-page: 1198
  issue: 475
  year: 2005
  ident: 10.1016/j.ejor.2017.03.051_bib0040
  article-title: PCA based on multivariate MM-estimators with fast and robust bootstrap
  publication-title: Journal of the American Statistical Association
  doi: 10.1198/016214506000000096
– volume: 67
  start-page: 301
  issue: 2
  year: 2005
  ident: 10.1016/j.ejor.2017.03.051_bib0046
  article-title: Regularization and variable selection via the elastic net
  publication-title: Journal of the Royal Statistical Society: Series B
  doi: 10.1111/j.1467-9868.2005.00503.x
– year: 2009
  ident: 10.1016/j.ejor.2017.03.051_bib0002
– year: 2009
  ident: 10.1016/j.ejor.2017.03.051_bib0027
– year: 2011
  ident: 10.1016/j.ejor.2017.03.051_bib0001
– volume: 1
  start-page: 799
  year: 1973
  ident: 10.1016/j.ejor.2017.03.051_bib0026
  article-title: Robust regression: Asymptotics, conjectures and Monte Carlo
  publication-title: The Annals of Statistics
  doi: 10.1214/aos/1176342503
– year: 1975
  ident: 10.1016/j.ejor.2017.03.051_bib0033
  article-title: On some topics in robustness
– volume: 79
  start-page: 871
  year: 1984
  ident: 10.1016/j.ejor.2017.03.051_bib0038
  article-title: Least median of squares regression
  publication-title: Journal of the American Statistical Association
  doi: 10.1080/01621459.1984.10477105
– volume: 42
  start-page: 324
  issue: 1
  year: 2014
  ident: 10.1016/j.ejor.2017.03.051_bib0016
  article-title: Adaptive robust variable selection
  publication-title: The Annals of Statistics
  doi: 10.1214/13-AOS1191
– year: 1977
  ident: 10.1016/j.ejor.2017.03.051_bib0024
– volume: 87
  start-page: 2283
  year: 2007
  ident: 10.1016/j.ejor.2017.03.051_bib0034
  article-title: Overview of total least-squares methods
  publication-title: Signal Processing
  doi: 10.1016/j.sigpro.2007.04.004
– start-page: 418
  year: 2010
  ident: 10.1016/j.ejor.2017.03.051_bib0036
  article-title: Solving structured sparsity regularization with proximal methods
– volume: 58
  start-page: 267
  year: 1996
  ident: 10.1016/j.ejor.2017.03.051_bib0043
  article-title: Regression shrinkage and selection via the Lasso
  publication-title: Journal of the Royal Statistical Society, Series B
  doi: 10.1111/j.2517-6161.1996.tb02080.x
– volume: 17
  start-page: 883
  issue: 6
  year: 1980
  ident: 10.1016/j.ejor.2017.03.051_bib0019
  article-title: An analysis of the total least squares problem
  publication-title: SIAM Journal of Numerical Analysis
  doi: 10.1137/0717073
– volume: 95
  start-page: 206
  year: 2005
  ident: 10.1016/j.ejor.2017.03.051_bib0013
  article-title: High breakdown estimators for principal components: The projection-pursuit approach revisited
  publication-title: Journal of Multivariate Analysis
  doi: 10.1016/j.jmva.2004.08.002
– volume: 18
  start-page: 1035
  issue: 4
  year: 1997
  ident: 10.1016/j.ejor.2017.03.051_bib0018
  article-title: Robust solutions to least-squares problems with uncertain data
  publication-title: SIAM Journal of Matrix Analysis and Applications
  doi: 10.1137/S0895479896298130
– ident: 10.1016/j.ejor.2017.03.051_bib0005
– volume: 1
  year: 1936
  ident: 10.1016/j.ejor.2017.03.051_sbref0014
  article-title: The approximation of one matrix by another of lower rank
  publication-title: Psychometrika
  doi: 10.1007/BF02288367
– year: 2004
  ident: 10.1016/j.ejor.2017.03.051_bib0007
– ident: 10.1016/j.ejor.2017.03.051_bib0041
– volume: 133
  start-page: 315
  year: 2005
  ident: 10.1016/j.ejor.2017.03.051_bib0030
  article-title: Consistency of the structured total least squares estimator in a multivariate errors-in-variables model
  publication-title: Journal of Statistical Planning and Inference
  doi: 10.1016/j.jspi.2003.12.020
– volume: 23
  start-page: 92
  issue: 1
  year: 2008
  ident: 10.1016/j.ejor.2017.03.051_bib0028
  article-title: High-breakdown robust multivariate methods
  publication-title: Statistical Science
  doi: 10.1214/088342307000000087
– volume: 69
  start-page: 383
  year: 1974
  ident: 10.1016/j.ejor.2017.03.051_bib0022
  article-title: The influence curve and its role in robust estimation
  publication-title: Journal of the American Statistical Association
  doi: 10.1080/01621459.1974.10482962
– year: 2007
  ident: 10.1016/j.ejor.2017.03.051_bib0042
  article-title: Soft modelling by latent variables: The nonlinear iterative partial least squares (NIPALS) approach
– volume: 48
  start-page: 3080
  issue: 5
  year: 2009
  ident: 10.1016/j.ejor.2017.03.051_bib0032
  article-title: Lipschitz behavior of the robust regularization
  publication-title: SIAM Journal on Control and Optimization
  doi: 10.1137/08073682X
– volume: 63
  start-page: 628
  issue: 3
  year: 2015
  ident: 10.1016/j.ejor.2017.03.051_bib0003
  article-title: Oracle-based robust optimization via online learning
  publication-title: Operations Research
  doi: 10.1287/opre.2015.1374
– volume: 56
  start-page: 3561
  issue: 7
  year: 2010
  ident: 10.1016/j.ejor.2017.03.051_bib0045
  article-title: Robust regression and Lasso
  publication-title: IEEE Transactions in Information Theory
  doi: 10.1109/TIT.2010.2048503
SSID ssj0001515
Score 2.570896
Snippet •We characterize the connection between robust optimization and regularization.•We extend the characterization to new settings such as matrix...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 931
SubjectTerms Adversarial learning
Convex programming
Penalty methods
Robust optimization
Statistical regression
Title Characterization of the equivalence of robustification and regularization in linear and matrix regression
URI https://dx.doi.org/10.1016/j.ejor.2017.03.051
Volume 270
WOSCitedRecordID wos000438660400013&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: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1872-6860
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001515
  issn: 0377-2217
  databaseCode: AIEXJ
  dateStart: 19950105
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3Pb9MwFH6qOoTYYYPCxMYP-cCtypQ4yZwcR9kECE1IDKm3yI5tkWpLR9ZN1f4G_mieY8fNCpvYgUsUuXbS5n2xP7--9z2Ad2XKo0hQGXBd8iBhLMazhAcprgaqDGOlRauu_4WdnGTTaf51MPjV5cJcn7G6zpbL_OK_mhrb0NgmdfYB5vYXxQY8R6PjEc2Ox38y_MRLMN94OthGbvy8qvA-7ZuMTc1cmEJe2vnsbJx5W5jeD6zqsSGh3IZZnhsx_6XpY0Nn6zud-o7gYkPTuRqdppD3Pb9XzeKyOrfZZB9MklVTeXY_MSW9fnAXW2qFDsbf9vsOiihzmXqreSxmLKDUpmh2ky615UIcuuLeFJq7VcGuxrnV3vpjorc-h9m-ms2NqmvEWqlap117S1V7bbXzMYhdeNusMNcozDWKMC5Ck4-_QXEblQ1h4_DT0fSzX9kN-Wv_lXI_yCVh2XjB9W_yd6LTIy-nT2HL7TrIoUXLMxioegSPu6SHEWx3xT2Im-tHsNlTqnwO1TqqyFwTRBXpoco0raGKIHDIbVSRqiYWVe2HFlVkhaoX8P346HTyMXBVOoIySegiSCQyGi1Ylkp8ZpTnTEZaHWQRkzxORRzKUisZc1UKoYQ2oayUHyjjakC-iH12YFjPa_USCE0Ez7Wp8MBEImjIQ4k77lAgzcddfSp3IeoeaVE6CXtTSeWsuNuYuzD2Yy6sgMu9vdPOUoWjoJZaFgi8e8btPegur-DJ6kV5DcNFc6XewKPyelFdNm8d6n4DMT6vRg
linkProvider Elsevier
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=Characterization+of+the+equivalence+of+robustification+and+regularization+in+linear+and+matrix+regression&rft.jtitle=European+journal+of+operational+research&rft.au=Bertsimas%2C+Dimitris&rft.au=Copenhaver%2C+Martin+S.&rft.date=2018-11-01&rft.issn=0377-2217&rft.volume=270&rft.issue=3&rft.spage=931&rft.epage=942&rft_id=info:doi/10.1016%2Fj.ejor.2017.03.051&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_ejor_2017_03_051
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0377-2217&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0377-2217&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0377-2217&client=summon