Multilayer Convolutional Sparse Modeling: Pursuit and Dictionary Learning

The recently proposed multilayer convolutional sparse coding (ML-CSC) model, consisting of a cascade of convolutional sparse layers, provides a new interpretation of convolutional neural networks (CNNs). Under this framework, the forward pass in a CNN is equivalent to a pursuit algorithm aiming to e...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on signal processing Jg. 66; H. 15; S. 4090 - 4104
Hauptverfasser: Sulam, Jeremias, Papyan, Vardan, Romano, Yaniv, Elad, Michael
Format: Journal Article
Sprache:Englisch
Veröffentlicht: IEEE 01.08.2018
Schlagworte:
ISSN:1053-587X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract The recently proposed multilayer convolutional sparse coding (ML-CSC) model, consisting of a cascade of convolutional sparse layers, provides a new interpretation of convolutional neural networks (CNNs). Under this framework, the forward pass in a CNN is equivalent to a pursuit algorithm aiming to estimate the nested sparse representation vectors from a given input signal. Despite having served as a pivotal connection between CNNs and sparse modeling, a deeper understanding of the ML-CSC is still lacking. In this paper, we propose a sound pursuit algorithm for the ML-CSC model by adopting a projection approach. We provide new and improved bounds on the stability of the solution of such pursuit and we analyze different practical alternatives to implement this in practice. We show that the training of the filters is essential to allow for nontrivial signals in the model, and we derive an online algorithm to learn the dictionaries from real data, effectively resulting in cascaded sparse convolutional layers. Last, but not least, we demonstrate the applicability of the ML-CSC model for several applications in an unsupervised setting, providing competitive results. Our work represents a bridge between matrix factorization, sparse dictionary learning, and sparse autoencoders, and we analyze these connections in detail.
AbstractList The recently proposed multilayer convolutional sparse coding (ML-CSC) model, consisting of a cascade of convolutional sparse layers, provides a new interpretation of convolutional neural networks (CNNs). Under this framework, the forward pass in a CNN is equivalent to a pursuit algorithm aiming to estimate the nested sparse representation vectors from a given input signal. Despite having served as a pivotal connection between CNNs and sparse modeling, a deeper understanding of the ML-CSC is still lacking. In this paper, we propose a sound pursuit algorithm for the ML-CSC model by adopting a projection approach. We provide new and improved bounds on the stability of the solution of such pursuit and we analyze different practical alternatives to implement this in practice. We show that the training of the filters is essential to allow for nontrivial signals in the model, and we derive an online algorithm to learn the dictionaries from real data, effectively resulting in cascaded sparse convolutional layers. Last, but not least, we demonstrate the applicability of the ML-CSC model for several applications in an unsupervised setting, providing competitive results. Our work represents a bridge between matrix factorization, sparse dictionary learning, and sparse autoencoders, and we analyze these connections in detail.
Author Elad, Michael
Romano, Yaniv
Papyan, Vardan
Sulam, Jeremias
Author_xml – sequence: 1
  givenname: Jeremias
  surname: Sulam
  fullname: Sulam, Jeremias
  email: jsulam@cs.technion.ac.il
  organization: Dept. of Comput. Sci., Technion - Israel Inst. of Technol., Haifa, Israel
– sequence: 2
  givenname: Vardan
  surname: Papyan
  fullname: Papyan, Vardan
  email: vardanp91@gmail.com
  organization: Dept. of Stat., Stanford Univ., Stanford, CA, USA
– sequence: 3
  givenname: Yaniv
  surname: Romano
  fullname: Romano, Yaniv
  email: yromano@tx.technion.ac.il
  organization: Dept. of Stat., Stanford Univ., Stanford, CA, USA
– sequence: 4
  givenname: Michael
  surname: Elad
  fullname: Elad, Michael
  email: elad@cs.technion.ac.il
  organization: Dept. of Comput. Sci., Technion - Israel Inst. of Technol., Haifa, Israel
BookMark eNotjstKAzEUQLOoYFvdC27yAzPm5jFJ3Mn4Kkyx0AruSiZzRyIxU-Yh9O-tj9XZHA5nQWapS0jIFbAcgNmb3XaTcwYm50YWnBczMgemRKaMfjsni2H4YAyktMWcrNZTHEN0R-xp2aWvLk5j6JKLdHtw_YB03TUYQ3q_pZupH6YwUpcaeh_8r9YfaYWuTyfhgpy1Lg54-c8leX182JXPWfXytCrvqiyAVmNW26JopUarpQDkQmlvpGVetdbWILhvrK8ZePhZbNEgE4a1DXdOgRIgxJJc_3UDIu4Pffg8XeyN0FYqJb4B8NJLeg
CODEN ITPRED
ContentType Journal Article
DBID 97E
RIA
RIE
DOI 10.1109/TSP.2018.2846226
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EndPage 4104
ExternalDocumentID 8379455
Genre orig-research
GrantInformation_xml – fundername: European Research Council
– fundername: European Unions Seventh Framework Programme ERC
  grantid: 320649
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
ID FETCH-LOGICAL-i175t-b966f47e97431e2357c8490c5f99b132cd9cb01c10014fe8e0380fd2aa5153133
IEDL.DBID RIE
ISICitedReferencesCount 74
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000436458100002&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 Wed Aug 27 02:48:28 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 15
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i175t-b966f47e97431e2357c8490c5f99b132cd9cb01c10014fe8e0380fd2aa5153133
PageCount 15
ParticipantIDs ieee_primary_8379455
PublicationCentury 2000
PublicationDate 2018-Aug.1,-1
PublicationDateYYYYMMDD 2018-08-01
PublicationDate_xml – month: 08
  year: 2018
  text: 2018-Aug.1,-1
  day: 01
PublicationDecade 2010
PublicationTitle IEEE transactions on signal processing
PublicationTitleAbbrev TSP
PublicationYear 2018
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0014496
Score 2.5832756
Snippet The recently proposed multilayer convolutional sparse coding (ML-CSC) model, consisting of a cascade of convolutional sparse layers, provides a new...
SourceID ieee
SourceType Publisher
StartPage 4090
SubjectTerms Adaptation models
Convolution
Convolutional codes
convolutional neural networks
Convolutional sparse coding
Data models
Dictionaries
dictionary learning
Machine learning
multilayer pursuit
Pursuit algorithms
sparse convolutional filters
Title Multilayer Convolutional Sparse Modeling: Pursuit and Dictionary Learning
URI https://ieeexplore.ieee.org/document/8379455
Volume 66
WOSCitedRecordID wos000436458100002&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)
  issn: 1053-587X
  databaseCode: RIE
  dateStart: 19910101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://ieeexplore.ieee.org/
  omitProxy: false
  ssIdentifier: ssj0014496
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV09a8MwEBVJ6NAO_UpLv9HQsUqUWLakbqVtaJcQSArZgqyPYihOcOxA_31PshsydOlmBAb7ZHzv6e69Q-heUaulZYo4ESeEcaqITKQh1KhIpZBwNVdh2AQfj8V8Lict9LDVwlhrQ_OZ7fnLUMs3S135o7I-kCnJ4riN2pwntVZrWzFgLMziArgQkVjw-W9Jksr-bDrxPVyiB7_iZOhtFHYGqYQ8Mjr63xMco8MGL-KneoNPUMvmp-hgx0Wwi96DiPZLAXjGz8t803xMcNd0BbTVYj_vzKvOH_GkKtZVVmKVG_ySBUmDKr5xY7L6eYY-Rq-z5zfSTEggGaT9kqRAVhzjVnocYL1zjRZMUh07KVPgmdpIndKB9kZLzFlhaSSoM0OlAMZEQE_PUSdf5vYCYUAeYqA5oFilgAUpAe-eDpwzIqZDaeQl6vpwLFa1CcaiicTV38vXaN9HvO6Uu0GdsqjsLdrTmzJbF3dh534AamOZIQ
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFA5zCurB3-Jvc_BotqxN18SbTMeGcww2YbeRJqkUpBtdO_C_9yWtYwcv3kqg0L6Uvu_Le9_3EHqQ1ChhmCQxD9qEhVQS0RaaUC19GUHCVaF0wybC4ZBPp2JUQ49rLYwxxjWfmYa9dLV8PVeFPSprApkSLAi20HbAmEdLtda6ZsCYm8YFgMEnAQ-nv0VJKpqT8ch2cfEG_IzbnjVS2Bil4jJJ9_B_z3CEDirEiJ_LLT5GNZOeoP0NH8FT1Hcy2i8J8Bl35umq-pzgrvECiKvBduKZ1Z0_4VGRLYskxzLV-CVxogaZfePKZvXzDH10XyedHqlmJJAEEn9OIqArMQuNsEjAWO8axZmgKoiFiIBpKi1URFvKWi2x2HBDfU5j7UkJQMYHgnqO6uk8NRcIA_bgLRUCjpUSeJDk8O5RK441D6gntLhEpzYcs0VpgzGrInH19_I92u1N3gezQX_4do32bPTLvrkbVM-zwtyiHbXKk2V253bxB5O1nGg
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=Multilayer+Convolutional+Sparse+Modeling%3A+Pursuit+and+Dictionary+Learning&rft.jtitle=IEEE+transactions+on+signal+processing&rft.au=Sulam%2C+Jeremias&rft.au=Papyan%2C+Vardan&rft.au=Romano%2C+Yaniv&rft.au=Elad%2C+Michael&rft.date=2018-08-01&rft.pub=IEEE&rft.issn=1053-587X&rft.volume=66&rft.issue=15&rft.spage=4090&rft.epage=4104&rft_id=info:doi/10.1109%2FTSP.2018.2846226&rft.externalDocID=8379455
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