L2/3 regularization: Convergence of iterative thresholding algorithm

•Under some condition, the sequence generated by the L2/3 algorithm converges to a local minimizer of L2/3 regularization.•Under the same conditions, the asymptotical convergence rate of L2/3 algorithm is linear.•Numerical experiments support our theoretical analysis. The L2/3 regularization is a no...

Full description

Saved in:
Bibliographic Details
Published in:Journal of visual communication and image representation Vol. 33; pp. 350 - 357
Main Authors: Zhang, Yong, Ye, Wanzhou
Format: Journal Article
Language:English
Published: Elsevier Inc 01.11.2015
Subjects:
ISSN:1047-3203, 1095-9076
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract •Under some condition, the sequence generated by the L2/3 algorithm converges to a local minimizer of L2/3 regularization.•Under the same conditions, the asymptotical convergence rate of L2/3 algorithm is linear.•Numerical experiments support our theoretical analysis. The L2/3 regularization is a nonconvex and nonsmooth optimization problem. Cao et al. (2013) investigated that the L2/3 regularization is more effective in imaging deconvolution. The convergence issue of the iterative thresholding algorithm of L2/3 regularization problem (the L2/3 algorithm) hasn’t been addressed in Cao et al. (2013). In this paper, we study the convergence of the L2/3 algorithm. As the main result, we show that under certain conditions, the sequence {x(n)} generated by the L2/3 algorithm converges to a local minimizer of L2/3 regularization, and its asymptotical convergence rate is linear. We provide a set of experiments to verify our theoretical assertions and show the performance of the algorithm on sparse signal recovery. The established results provide a theoretical guarantee for a wide range of applications of the algorithm.
AbstractList •Under some condition, the sequence generated by the L2/3 algorithm converges to a local minimizer of L2/3 regularization.•Under the same conditions, the asymptotical convergence rate of L2/3 algorithm is linear.•Numerical experiments support our theoretical analysis. The L2/3 regularization is a nonconvex and nonsmooth optimization problem. Cao et al. (2013) investigated that the L2/3 regularization is more effective in imaging deconvolution. The convergence issue of the iterative thresholding algorithm of L2/3 regularization problem (the L2/3 algorithm) hasn’t been addressed in Cao et al. (2013). In this paper, we study the convergence of the L2/3 algorithm. As the main result, we show that under certain conditions, the sequence {x(n)} generated by the L2/3 algorithm converges to a local minimizer of L2/3 regularization, and its asymptotical convergence rate is linear. We provide a set of experiments to verify our theoretical assertions and show the performance of the algorithm on sparse signal recovery. The established results provide a theoretical guarantee for a wide range of applications of the algorithm.
Author Ye, Wanzhou
Zhang, Yong
Author_xml – sequence: 1
  givenname: Yong
  surname: Zhang
  fullname: Zhang, Yong
  email: 13820161@shu.edu.cn
– sequence: 2
  givenname: Wanzhou
  surname: Ye
  fullname: Ye, Wanzhou
  email: wzhy@shu.edu.cn
BookMark eNotkEFPhDAQhRuzJu6u_gIv_AHYaUsLNfFg0FUTEi96bkoZoARpUpCDv15QT2_y3mTe5DuQ3ehHJOSWQkKBylOf9It1IWFAxeokANkF2VNQIlaQyd02p1nMGfArcpimHgC44umePJbsxKOA7ddggvs2s_PjXVT4ccHQ4mgx8k3kZgxrsmA0dwGnzg-1G9vIDK0Pbu4-r8llY4YJb_71SD7OT-_FS1y-Pb8WD2WMlKVzXDPBci4MU9LmxqrMgEkVCtGkmaqrilKToYEqFbziOVNWCqkUXxexolJKfiT3f3dxLVkcBj1Ztz1Zu4B21rV3moLekOhe_yLRG5LNXJHwH5KoWMY
ContentType Journal Article
Copyright 2015 Elsevier Inc.
Copyright_xml – notice: 2015 Elsevier Inc.
DOI 10.1016/j.jvcir.2015.10.007
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Journalism & Communications
Engineering
EISSN 1095-9076
EndPage 357
ExternalDocumentID S1047320315001984
GroupedDBID --K
--M
.DC
.~1
0R~
1B1
1~.
1~5
29L
4.4
457
4G.
53G
5GY
5VS
7-5
71M
8P~
9JN
AACTN
AAEDT
AAEDW
AAIAV
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AAXUO
AAYFN
ABBOA
ABFNM
ABJNI
ABMAC
ABXDB
ABYKQ
ACDAQ
ACGFS
ACNNM
ACRLP
ACZNC
ADBBV
ADEZE
ADFGL
ADJOM
ADMHC
ADMUD
ADTZH
AEBSH
AECPX
AEKER
AENEX
AFKWA
AFTJW
AGHFR
AGUBO
AGYEJ
AHHHB
AHJVU
AHZHX
AIALX
AIEXJ
AIKHN
AITUG
AJBFU
AJOXV
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
AOUOD
ASPBG
AVWKF
AXJTR
AZFZN
BJAXD
BKOJK
BLXMC
CAG
COF
CS3
DM4
DU5
EBS
EFBJH
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-2
G-Q
GBLVA
GBOLZ
HLZ
HVGLF
HZ~
IHE
J1W
JJJVA
KOM
LG5
LX9
M41
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
R2-
RIG
ROL
RPZ
SBC
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SST
SSV
SSZ
T5K
WH7
WUQ
XPP
YQT
ZMT
ZU3
~G-
ID FETCH-LOGICAL-e124t-d252835a296c8ac97a0a49e55f479dbb11a7ea0b453b3829c656993c97eb16663
ISSN 1047-3203
IngestDate Fri Feb 23 02:24:19 EST 2024
IsPeerReviewed true
IsScholarly true
Keywords L2/3 regularization
Local minimizer
Iterative thresholding algorithm
L1/2 regularization
Thresholding formula
Sparse signal recovery
Asymptotical convergence rate
Convergence
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-e124t-d252835a296c8ac97a0a49e55f479dbb11a7ea0b453b3829c656993c97eb16663
PageCount 8
ParticipantIDs elsevier_sciencedirect_doi_10_1016_j_jvcir_2015_10_007
PublicationCentury 2000
PublicationDate November 2015
PublicationDateYYYYMMDD 2015-11-01
PublicationDate_xml – month: 11
  year: 2015
  text: November 2015
PublicationDecade 2010
PublicationTitle Journal of visual communication and image representation
PublicationYear 2015
Publisher Elsevier Inc
Publisher_xml – name: Elsevier Inc
References Donoho (b0085) 1995; 41
Cao, Sun, Xu (b0075) 2013; 24
F.P. Nie, H. Huang, C. Ding, Low-Rank matrix recovery via efficient Schatten
norm minimization, in: Proc. Twenty-Sixth AAAI Conference on Artificial Intelligence, 2012, pp. 655–661.
Candes, Romberg, Tao (b0050) 2006; 52
Rudelson, Vershynin (b0130) 2008; 61
Nie, Wang, Huang, Ding (b0060) 2015; 42
Baraniuk, Davenport, Devore, Wakin (b0135) 2008; 28
Foucart, Lai (b0115) 2009; 26
Donoho (b0045) 2006; 52
Zeng, Lin, Wang, Xu (b0105) 2014; 62
Tibshirani (b0025) 1996; 58
norms minimization, in: Proc. Adv. Neural Inf. Process. Syst. (NIPS), 2010, pp. 1813–1821.
Xu, Zhang, Wang, Chang (b0015) 2010; 53
H. Tao, C.P. Hou, F.P. Nie, Y.Y. Jiao, D.Y. Yi, Effective discriminative feature selection with non-trivial solutions, 2015. Available from
Gui, Tao, Sun, Luo, You, Tang (b0095) 2014; 23
Chartrand, Staneva (b0010) 2008; 24
Xu, Chang, Xu, Zhang (b0020) 2012; 23
T.L. Liu, D.C. Tao, On the performance of manhattan nonnegative matrix factorization, 2015.
Chartrand (b0005) 2007; 14
Liu, Chen (b0110) 2015; 32
D. Krishnan, R. Fergus, Fast image deconvolution using hyper-Laplacian priors, in: Proc. Adv. Neural Inf. Process. Syst. (NIPS), 2009.
F.P. Nie, H. Huang, X. Cai, C. Ding, Efficient and robust feature selection via joint
.
Daubechies, Defrise, Christine (b0090) 2004; 57
Tian, Tao, Rui (b0100) 2012; 8
Candès, Wakin, Boyd (b0055) 2008; 14
Sun, Liu, Tang, Tao (b0125) 2014; 23
Zhou, Tao (b0040) 2013; 22
Blumensath, Yaghoobi, Davies (b0080) 2007; 3
Sun (b0120) 2012; 32
References_xml – volume: 57
  start-page: 1413
  year: 2004
  end-page: 1457
  ident: b0090
  article-title: An iterative thresholding algorithm for linear inverse problems with a sparsity constraint
  publication-title: Commun. Pure Appl. Math.
– reference: -norms minimization, in: Proc. Adv. Neural Inf. Process. Syst. (NIPS), 2010, pp. 1813–1821.
– volume: 61
  start-page: 1025
  year: 2008
  end-page: 1045
  ident: b0130
  article-title: On sparse reconstruction from Fourier and Gaussian measurements
  publication-title: Comm. Pure Appl. Math.
– volume: 3
  year: 2007
  ident: b0080
  article-title: Iterative hard thresholding and
  publication-title: IEEE Trans. Acoust. Speech Signal Process.
– volume: 8
  start-page: 26:1
  year: 2012
  end-page: 19
  ident: b0100
  article-title: Sparse transfer learning for interactive video search reranking
  publication-title: ACM Trans. Multim. Comput. Appl.
– volume: 32
  start-page: 329
  year: 2012
  end-page: 341
  ident: b0120
  article-title: Recovery of sparsest signals via
  publication-title: Appl. Comput. Harmonic Anal.
– volume: 23
  start-page: 3816
  year: 2014
  end-page: 3828
  ident: b0125
  article-title: Learning discriminative dictionary for group sparse representation
  publication-title: IEEE Trans. Image Process.
– volume: 32
  start-page: 1550023
  year: 2015
  ident: b0110
  article-title: Convergence of
  publication-title: Asia-Pacific J. Oper. Res.
– reference: D. Krishnan, R. Fergus, Fast image deconvolution using hyper-Laplacian priors, in: Proc. Adv. Neural Inf. Process. Syst. (NIPS), 2009.
– reference: F.P. Nie, H. Huang, X. Cai, C. Ding, Efficient and robust feature selection via joint
– reference: H. Tao, C.P. Hou, F.P. Nie, Y.Y. Jiao, D.Y. Yi, Effective discriminative feature selection with non-trivial solutions, 2015. Available from:
– volume: 22
  start-page: 244
  year: 2013
  end-page: 257
  ident: b0040
  article-title: Double shrinking sparse dimension reduction
  publication-title: IEEE Trans. Image Process.
– reference: -norm minimization, in: Proc. Twenty-Sixth AAAI Conference on Artificial Intelligence, 2012, pp. 655–661.
– volume: 52
  start-page: 1289
  year: 2006
  end-page: 1306
  ident: b0045
  article-title: Compressed sensing
  publication-title: IEEE Trans. Inf. Theory
– volume: 52
  start-page: 489
  year: 2006
  end-page: 509
  ident: b0050
  article-title: Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
  publication-title: IEEE Trans. Inf. Theory
– volume: 26
  start-page: 395
  year: 2009
  end-page: 407
  ident: b0115
  article-title: Sparsest solutions of underdetermined linear systems via
  publication-title: Appl. Comput. Harmonic Anal.
– volume: 23
  start-page: 1013
  year: 2012
  end-page: 1027
  ident: b0020
  article-title: regularization: a thresholding representation theory and a fast solver
  publication-title: IEEE Trans. Neural Netw. Learning Syst.
– volume: 62
  start-page: 2317
  year: 2014
  end-page: 2329
  ident: b0105
  article-title: regularization: convergence of iterative half thresholding algorithm
  publication-title: IEEE Trans. Signal Process.
– volume: 23
  start-page: 3126
  year: 2014
  end-page: 3137
  ident: b0095
  article-title: Group sparse multiview patch alignment framework with view consistency for image classification
  publication-title: IEEE Trans. Image Process.
– reference: F.P. Nie, H. Huang, C. Ding, Low-Rank matrix recovery via efficient Schatten
– volume: 28
  start-page: 253
  year: 2008
  end-page: 263
  ident: b0135
  article-title: A simple proof of the restricted isometry property for random matrices
  publication-title: Constr. Approx.
– volume: 53
  start-page: 1159
  year: 2010
  end-page: 1169
  ident: b0015
  article-title: regularization
  publication-title: Sci. China
– volume: 42
  start-page: 525
  year: 2015
  end-page: 544
  ident: b0060
  article-title: Joint Schatten
  publication-title: Knowl. Inf. Syst.
– reference: >.
– reference: .
– volume: 41
  start-page: 613
  year: 1995
  end-page: 627
  ident: b0085
  article-title: Denoising by soft thresholding
  publication-title: IEEE Trans. Inf. Theory
– volume: 14
  start-page: 877
  year: 2008
  end-page: 905
  ident: b0055
  article-title: Enhancing sparsity by reweighted
  publication-title: J. Fourier Anal. Appl.
– volume: 24
  start-page: 1
  year: 2008
  end-page: 14
  ident: b0010
  article-title: Restricted isometry properties and nonconvex compressive sensing
  publication-title: Inverse Prob.
– volume: 24
  start-page: 1529
  year: 2013
  end-page: 1542
  ident: b0075
  article-title: Fast image deconvolution using closed-form thresholding formulas of
  publication-title: J. Vis. Commun. Image Represent.
– volume: 58
  start-page: 267
  year: 1996
  end-page: 288
  ident: b0025
  article-title: Regression shrinkage and selection via the lasso
  publication-title: J. Roy. Statist. Soc., Ser. B (Methodolog.)
– reference: T.L. Liu, D.C. Tao, On the performance of manhattan nonnegative matrix factorization, 2015. <
– volume: 14
  start-page: 707
  year: 2007
  end-page: 710
  ident: b0005
  article-title: Exact reconstruction of sparse signals via nonconvex minimization
  publication-title: IEEE Signal Process. Lett.
SSID ssj0003934
Score 2.1767178
Snippet •Under some condition, the sequence generated by the L2/3 algorithm converges to a local minimizer of L2/3 regularization.•Under the same conditions, the...
SourceID elsevier
SourceType Publisher
StartPage 350
SubjectTerms [formula omitted] regularization
Asymptotical convergence rate
Convergence
Iterative thresholding algorithm
Local minimizer
Sparse signal recovery
Thresholding formula
Title L2/3 regularization: Convergence of iterative thresholding algorithm
URI https://dx.doi.org/10.1016/j.jvcir.2015.10.007
Volume 33
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1095-9076
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0003934
  issn: 1047-3203
  databaseCode: AIEXJ
  dateStart: 19950301
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3PS9xAFB62q4d6KNZaarUyB_Ei0U0mycx4E6tYWUSoLespTJJJzeImJcku_vl98yMxiwgq9BKWYTYZ5nu892Xy3vcQ2qMQI4GIB-D9Mub4goWOIAlxYtVdJKRukuks399jenXFJhN-PRhkbS3M4p4WBXt44H__K9QwBmCr0tlXwN3dFAbgN4AOV4Adri8Cfqy0TslBpZvMV7bMUr34n6oE88qIb2qdWWlVvxvAs7afoQ7E_Z-yypu72TO0dZHXc60p0issMV8gZir9R4tktgVNneF1x9K3pY2U-phWJ_gBOb0r5_3TBzewZXjdkVhbFrOUtanlH4g3Mp5L2jHVFHJkmr20rpeQnu8kRoHWhmFidKufeHhz2DA9nC6SXOm5usGhzs6jjwGtSzP8qRai1gGsF6gs89-hFY8GnA3RysmPs8llF7MJN_kH7cJbfSqdCfjkUT3i0iMjN-vog4UDnxj0P6KBLDbQWk9bcgNt20l5PcP7eKkOqP6Evo-9I4KXbeQY9ywElxnuLAT3LQR3FrKJfp2f3ZxeOLahhiOBxjVO6ikpn0B4PEyYSDgVI-FzGQSZT3kax64rqBSj2A9ITJjHEyD7wF9hIkR0eM8ln9GwKAv5BWEKM0kqKOwQ89PUZdIXIhSpEmjkQIq3UNhuUWS5nOFoEQAZtamF00jvbaT2Vg3C3n596x-30ftHA91Bw6aay29oNVk0eV3tWsD_AXbYb9o
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=L2%2F3+regularization%3A+Convergence+of+iterative+thresholding+algorithm&rft.jtitle=Journal+of+visual+communication+and+image+representation&rft.au=Zhang%2C+Yong&rft.au=Ye%2C+Wanzhou&rft.date=2015-11-01&rft.pub=Elsevier+Inc&rft.issn=1047-3203&rft.eissn=1095-9076&rft.volume=33&rft.spage=350&rft.epage=357&rft_id=info:doi/10.1016%2Fj.jvcir.2015.10.007&rft.externalDocID=S1047320315001984
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1047-3203&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1047-3203&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1047-3203&client=summon