Disentangling Representations in Restricted Boltzmann Machines without Adversaries

A goal of unsupervised machine learning is to build representations of complex high-dimensional data, with simple relations to their properties. Such disentangled representations make it easier to interpret the significant latent factors of variation in the data, as well as to generate new data with...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Physical review. X Ročník 13; číslo 2; s. 021003
Hlavní autori: Fernandez-de-Cossio-Diaz, Jorge, Cocco, Simona, Monasson, Rémi
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: College Park American Physical Society 01.04.2023
Predmet:
ISSN:2160-3308, 2160-3308
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract A goal of unsupervised machine learning is to build representations of complex high-dimensional data, with simple relations to their properties. Such disentangled representations make it easier to interpret the significant latent factors of variation in the data, as well as to generate new data with desirable features. The methods for disentangling representations often rely on an adversarial scheme, in which representations are tuned to avoid discriminators from being able to reconstruct information about the data properties (labels). Unfortunately, adversarial training is generally difficult to implement in practice. Here we propose a simple, effective way of disentangling representations without any need to train adversarial discriminators and apply our approach to Restricted Boltzmann Machines, one of the simplest representation-based generative models. Our approach relies on the introduction of adequate constraints on the weights during training, which allows us to concentrate information about labels on a small subset of latent variables. The effectiveness of the approach is illustrated with four examples: the CelebA dataset of facial images, the two-dimensional Ising model, the MNIST dataset of handwritten digits, and the taxonomy of protein families. In addition, we show how our framework allows for analytically computing the cost, in terms of the log-likelihood of the data, associated with the disentanglement of their representations.
AbstractList A goal of unsupervised machine learning is to build representations of complex high-dimensional data, with simple relations to their properties. Such disentangled representations make it easier to interpret the significant latent factors of variation in the data, as well as to generate new data with desirable features. The methods for disentangling representations often rely on an adversarial scheme, in which representations are tuned to avoid discriminators from being able to reconstruct information about the data properties (labels). Unfortunately, adversarial training is generally difficult to implement in practice. Here we propose a simple, effective way of disentangling representations without any need to train adversarial discriminators and apply our approach to Restricted Boltzmann Machines, one of the simplest representation-based generative models. Our approach relies on the introduction of adequate constraints on the weights during training, which allows us to concentrate information about labels on a small subset of latent variables. The effectiveness of the approach is illustrated with four examples: the CelebA dataset of facial images, the two-dimensional Ising model, the MNIST dataset of handwritten digits, and the taxonomy of protein families. In addition, we show how our framework allows for analytically computing the cost, in terms of the log-likelihood of the data, associated with the disentanglement of their representations.
ArticleNumber 021003
Author Monasson, Rémi
Fernandez-de-Cossio-Diaz, Jorge
Cocco, Simona
Author_xml – sequence: 1
  givenname: Jorge
  orcidid: 0000-0002-4476-805X
  surname: Fernandez-de-Cossio-Diaz
  fullname: Fernandez-de-Cossio-Diaz, Jorge
– sequence: 2
  givenname: Simona
  orcidid: 0000-0002-1852-7789
  surname: Cocco
  fullname: Cocco, Simona
– sequence: 3
  givenname: Rémi
  orcidid: 0000-0002-4459-0204
  surname: Monasson
  fullname: Monasson, Rémi
BackLink https://hal.science/hal-04307038$$DView record in HAL
BookMark eNp9kUFvEzEQhS1UJNrSP8BppZ44JIzt3bX3GNpCKwWBolbiZk1sb-Joa6e2E1R-PU4XUOmBuXj09N4njd8JOfLBW0LeUZhSCvzDt_VjWtj99ynlU2AUgL8ix4y2MOEc5NGz_Q05S2kDZVqgtRDHZHHpkvUZ_WpwflUt7DbaJyG74FPlfJFSjk5na6qPYcg_79H76gvqtfM2VT9cXoddrmZmb2PC6Gx6S173OCR79vs9JXefrm4vrifzr59vLmbzia5plyeyMaib3nTcgmCa9Z1hWvayXda0Zx0VLYeGdgylwbaWKMuptVzWpte17FHzU3Izck3AjdpGd4_xUQV06kkIcaUwZqcHq5aiRd5QbXkBoC0wIbpWgoAGODO0sN6PrDUO_6CuZ3N10KDmxc3l_uA9H73bGB525XfUJuyiL6cqJjoJXcskKy42unQMKUXb_8VSUIfa1J_aFOVqrK2E5IuQdmMVOaIb_hf9BV-ooFA
CitedBy_id crossref_primary_10_1088_2632_2153_adf521
crossref_primary_10_1063_5_0226007
crossref_primary_10_1051_epn_2025109
crossref_primary_10_1088_1742_5468_ad292d
Cites_doi 10.1109/MSP.2012.2211477
10.1103/PhysRev.65.117
10.1103/PhysRevB.100.064304
10.1070/SM1967v001n04ABEH001994
10.1088/1361-6633/aa9965
10.1016/S0092-8674(00)81100-9
10.1017/CBO9781139164542
10.1017/CBO9780511810800
10.1140/epjb/e2006-00209-7
10.1038/nrm2178
10.1103/PhysRevE.99.032113
10.7554/eLife.39397
10.1103/PhysRevLett.118.138301
10.1146/annurev.neuro.25.112701.142909
10.1093/oso/9780198517962.001.0001
10.1103/PhysRevE.100.032128
10.1103/PhysRevResearch.2.043390
10.1073/pnas.1111471108
10.1093/nar/gky995
10.1093/nar/29.3.638
10.21468/SciPostPhys.8.5.074
10.1371/journal.pcbi.1007544
10.7551/mitpress/4175.001.0001
10.1093/nar/gkab314
10.1016/S0896-6273(00)81205-2
10.1038/s41587-022-01307-0
10.1093/nar/gkaa1100
10.1088/1742-5468/ac98a7
10.1016/j.cels.2020.11.005
10.1111/j.1742-4658.2008.06411.x
10.1038/s41592-022-01488-1
10.1103/RevModPhys.77.579
10.1109/TIP.2019.2916751
10.1016/j.cpc.2020.107518
10.1088/1751-8121/ab7d00
ContentType Journal Article
Copyright 2023. This work is licensed under https://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Attribution
Copyright_xml – notice: 2023. This work is licensed under https://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
– notice: Attribution
DBID AAYXX
CITATION
3V.
7XB
88I
8FE
8FG
8FK
ABJCF
ABUWG
AFKRA
AZQEC
BENPR
BGLVJ
CCPQU
DWQXO
GNUQQ
HCIFZ
L6V
M2P
M7S
PHGZM
PHGZT
PIMPY
PKEHL
PQEST
PQGLB
PQQKQ
PQUKI
PRINS
PTHSS
Q9U
1XC
VOOES
DOA
DOI 10.1103/PhysRevX.13.021003
DatabaseName CrossRef
ProQuest Central (Corporate)
ProQuest Central (purchase pre-March 2016)
Science Database (Alumni Edition)
ProQuest SciTech Collection
ProQuest Technology Collection
ProQuest Central (Alumni) (purchase pre-March 2016)
Materials Science & Engineering Collection
ProQuest Central (Alumni)
ProQuest Central UK/Ireland
ProQuest Central Essentials
ProQuest Central
Technology collection
ProQuest One Community College
ProQuest Central
ProQuest Central Student
SciTech Premium Collection
ProQuest Engineering Collection
Science Database
Engineering Database
ProQuest Central Premium
ProQuest One Academic
ProQuest Publicly Available Content
ProQuest One Academic Middle East (New)
ProQuest One Academic Eastern Edition (DO NOT USE)
ProQuest One Applied & Life Sciences
ProQuest One Academic (retired)
ProQuest One Academic UKI Edition
ProQuest Central China
Engineering Collection
ProQuest Central Basic
Hyper Article en Ligne (HAL)
Hyper Article en Ligne (HAL) (Open Access)
DOAJ Directory of Open Access Journals
DatabaseTitle CrossRef
Publicly Available Content Database
ProQuest Central Student
Technology Collection
ProQuest One Academic Middle East (New)
ProQuest Central Essentials
ProQuest Central (Alumni Edition)
SciTech Premium Collection
ProQuest One Community College
ProQuest Central China
ProQuest Central
ProQuest One Applied & Life Sciences
ProQuest Engineering Collection
ProQuest Central Korea
ProQuest Central (New)
Engineering Collection
Engineering Database
ProQuest Science Journals (Alumni Edition)
ProQuest Central Basic
ProQuest Science Journals
ProQuest One Academic Eastern Edition
ProQuest Technology Collection
ProQuest SciTech Collection
ProQuest One Academic UKI Edition
Materials Science & Engineering Collection
ProQuest One Academic
ProQuest One Academic (New)
ProQuest Central (Alumni)
DatabaseTitleList Publicly Available Content Database

Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: PIMPY
  name: ProQuest Publicly Available Content
  url: http://search.proquest.com/publiccontent
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Physics
Computer Science
EISSN 2160-3308
ExternalDocumentID oai_doaj_org_article_b76a351ce38b4ae8a8779680705032d1
oai:HAL:hal-04307038v1
10_1103_PhysRevX_13_021003
GroupedDBID 3MX
5VS
88I
AAFWJ
AAYXX
ABJCF
ABSSX
ABUWG
ADBBV
AECSF
AENEX
AFFHD
AFGMR
AFKRA
AFPKN
AGDNE
ALMA_UNASSIGNED_HOLDINGS
AUAIK
AZQEC
BCNDV
BENPR
BGLVJ
CCPQU
CITATION
DWQXO
EBS
EJD
FRP
GNUQQ
GROUPED_DOAJ
HCIFZ
KQ8
M2P
M7S
M~E
OK1
PHGZM
PHGZT
PIMPY
PQGLB
PTHSS
ROL
S7W
3V.
7XB
8FE
8FG
8FK
L6V
PKEHL
PQEST
PQQKQ
PQUKI
PRINS
Q9U
1XC
VOOES
ID FETCH-LOGICAL-c419t-85dac5fd93e072c2f9d2c8f86b41f29176305192a8da648a810348b4dfc48fac3
IEDL.DBID BENPR
ISICitedReferencesCount 13
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000995916500001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2160-3308
IngestDate Fri Oct 03 12:44:55 EDT 2025
Sat Nov 01 11:26:59 EDT 2025
Fri Jul 25 11:52:16 EDT 2025
Sat Nov 29 02:51:07 EST 2025
Tue Nov 18 21:02:29 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 2
Language English
License Attribution: http://creativecommons.org/licenses/by
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c419t-85dac5fd93e072c2f9d2c8f86b41f29176305192a8da648a810348b4dfc48fac3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ORCID 0000-0002-4476-805X
0000-0002-4459-0204
0000-0002-1852-7789
OpenAccessLink https://www.proquest.com/docview/2798096282?pq-origsite=%requestingapplication%
PQID 2798096282
PQPubID 5161131
ParticipantIDs doaj_primary_oai_doaj_org_article_b76a351ce38b4ae8a8779680705032d1
hal_primary_oai_HAL_hal_04307038v1
proquest_journals_2798096282
crossref_primary_10_1103_PhysRevX_13_021003
crossref_citationtrail_10_1103_PhysRevX_13_021003
PublicationCentury 2000
PublicationDate 2023-04-01
PublicationDateYYYYMMDD 2023-04-01
PublicationDate_xml – month: 04
  year: 2023
  text: 2023-04-01
  day: 01
PublicationDecade 2020
PublicationPlace College Park
PublicationPlace_xml – name: College Park
PublicationTitle Physical review. X
PublicationYear 2023
Publisher American Physical Society
Publisher_xml – name: American Physical Society
References PhysRevX.13.021003Cc20R1
PhysRevX.13.021003Cc43R1
PhysRevX.13.021003Cc41R1
PhysRevX.13.021003Cc60R1
PhysRevX.13.021003Cc24R1
PhysRevX.13.021003Cc47R1
PhysRevX.13.021003Cc45R1
M. E. Newman (PhysRevX.13.021003Cc37R1) 1999
PhysRevX.13.021003Cc28R1
PhysRevX.13.021003Cc49R1
Y. Shen (PhysRevX.13.021003Cc12R1) 2020
Z. Liu (PhysRevX.13.021003Cc23R1) 2015
PhysRevX.13.021003Cc50R1
B. Scholkopf (PhysRevX.13.021003Cc33R1) 2018
PhysRevX.13.021003Cc35R1
Q. Hu (PhysRevX.13.021003Cc9R1) 2018
S. Goldt (PhysRevX.13.021003Cc55R1) 2022
PhysRevX.13.021003Cc5R1
PhysRevX.13.021003Cc39R1
PhysRevX.13.021003Cc58R1
I. Higgins (PhysRevX.13.021003Cc54R1) 2017
B. Esmaeili (PhysRevX.13.021003Cc10R1) 2019
G. E. Hinton (PhysRevX.13.021003Cc18R1) 2012
K. M. Abadir (PhysRevX.13.021003Cc22R1) 2005
A. Engel (PhysRevX.13.021003Cc31R1) 2001
PhysRevX.13.021003Cc42R1
PhysRevX.13.021003Cc40R1
PhysRevX.13.021003Cc61R1
R. Salakhutdinov (PhysRevX.13.021003Cc21R1) 2007
PhysRevX.13.021003Cc25R1
PhysRevX.13.021003Cc46R1
PhysRevX.13.021003Cc44R1
PhysRevX.13.021003Cc29R1
PhysRevX.13.021003Cc48R1
R. Zemel (PhysRevX.13.021003Cc15R1) 2013
PhysRevX.13.021003Cc32R1
T. Tieleman (PhysRevX.13.021003Cc56R1) 2008
H. Kim (PhysRevX.13.021003Cc8R1) 2018
T. M. Cover (PhysRevX.13.021003Cc30R1) 1999
PhysRevX.13.021003Cc51R1
Y. Bengio (PhysRevX.13.021003Cc1R1) 2012
PhysRevX.13.021003Cc36R1
Y. Burda (PhysRevX.13.021003Cc53R1) 2015
PhysRevX.13.021003Cc59R1
R. Salakhutdinov (PhysRevX.13.021003Cc2R1) 2009
F. Locatello (PhysRevX.13.021003Cc6R1) 2019
PhysRevX.13.021003Cc11R1
PhysRevX.13.021003Cc34R1
PhysRevX.13.021003Cc38R1
PhysRevX.13.021003Cc57R1
PhysRevX.13.021003Cc19R1
R. J. Baxter (PhysRevX.13.021003Cc27R1) 2016
References_xml – volume-title: Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics
  year: 2019
  ident: PhysRevX.13.021003Cc10R1
– ident: PhysRevX.13.021003Cc24R1
  doi: 10.1109/MSP.2012.2211477
– volume-title: Proceedings of the 24th International Conference on Machine Learning
  year: 2007
  ident: PhysRevX.13.021003Cc21R1
– ident: PhysRevX.13.021003Cc36R1
  doi: 10.1103/PhysRev.65.117
– volume-title: Exactly Solved Models in Statistical Mechanics
  year: 2016
  ident: PhysRevX.13.021003Cc27R1
– ident: PhysRevX.13.021003Cc39R1
  doi: 10.1103/PhysRevB.100.064304
– volume-title: Proceedings of the 36th International Conference on Machine Learning
  year: 2019
  ident: PhysRevX.13.021003Cc6R1
– ident: PhysRevX.13.021003Cc34R1
  doi: 10.1070/SM1967v001n04ABEH001994
– ident: PhysRevX.13.021003Cc49R1
  doi: 10.1088/1361-6633/aa9965
– ident: PhysRevX.13.021003Cc45R1
  doi: 10.1016/S0092-8674(00)81100-9
– volume-title: Statistical Mechanics of Learning
  year: 2001
  ident: PhysRevX.13.021003Cc31R1
  doi: 10.1017/CBO9781139164542
– volume-title: Proceedings of the 18th International Conference on Artificial Intelligence and Statistics
  year: 2015
  ident: PhysRevX.13.021003Cc53R1
– volume-title: Matrix Algebra
  year: 2005
  ident: PhysRevX.13.021003Cc22R1
  doi: 10.1017/CBO9780511810800
– volume-title: Proceedings of International Conference on Computer Vision
  year: 2015
  ident: PhysRevX.13.021003Cc23R1
– ident: PhysRevX.13.021003Cc40R1
  doi: 10.1140/epjb/e2006-00209-7
– ident: PhysRevX.13.021003Cc43R1
  doi: 10.1038/nrm2178
– ident: PhysRevX.13.021003Cc38R1
  doi: 10.1103/PhysRevE.99.032113
– ident: PhysRevX.13.021003Cc19R1
  doi: 10.7554/eLife.39397
– ident: PhysRevX.13.021003Cc61R1
  doi: 10.1103/PhysRevLett.118.138301
– volume-title: Proceedings of the 35th International Conference on Machine Learning
  year: 2018
  ident: PhysRevX.13.021003Cc8R1
– ident: PhysRevX.13.021003Cc46R1
  doi: 10.1146/annurev.neuro.25.112701.142909
– volume-title: Monte Carlo Methods in Statistical Physics
  year: 1999
  ident: PhysRevX.13.021003Cc37R1
  doi: 10.1093/oso/9780198517962.001.0001
– ident: PhysRevX.13.021003Cc41R1
  doi: 10.1103/PhysRevE.100.032128
– ident: PhysRevX.13.021003Cc58R1
  doi: 10.1103/PhysRevResearch.2.043390
– ident: PhysRevX.13.021003Cc50R1
  doi: 10.1073/pnas.1111471108
– ident: PhysRevX.13.021003Cc25R1
  doi: 10.1093/nar/gky995
– ident: PhysRevX.13.021003Cc42R1
  doi: 10.1093/nar/29.3.638
– ident: PhysRevX.13.021003Cc59R1
  doi: 10.21468/SciPostPhys.8.5.074
– volume-title: Proceedings of the ICML Workshop on Unsupervised and Transfer Learning
  year: 2012
  ident: PhysRevX.13.021003Cc1R1
– ident: PhysRevX.13.021003Cc5R1
  doi: 10.1371/journal.pcbi.1007544
– volume-title: Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond
  year: 2018
  ident: PhysRevX.13.021003Cc33R1
  doi: 10.7551/mitpress/4175.001.0001
– volume-title: Proceedings of the 5th International Conference on Learning Representations
  year: 2017
  ident: PhysRevX.13.021003Cc54R1
– ident: PhysRevX.13.021003Cc47R1
  doi: 10.1093/nar/gkab314
– ident: PhysRevX.13.021003Cc32R1
  doi: 10.1016/S0896-6273(00)81205-2
– volume-title: Proceedings of the 2nd Mathematical and Scientific Machine Learning Conference
  year: 2022
  ident: PhysRevX.13.021003Cc55R1
– volume-title: Proceedings of the 25th International Conference on Machine Learning
  year: 2008
  ident: PhysRevX.13.021003Cc56R1
– ident: PhysRevX.13.021003Cc60R1
  doi: 10.1038/s41587-022-01307-0
– ident: PhysRevX.13.021003Cc48R1
  doi: 10.1093/nar/gkaa1100
– volume-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition
  year: 2018
  ident: PhysRevX.13.021003Cc9R1
– volume-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition
  year: 2020
  ident: PhysRevX.13.021003Cc12R1
– volume-title: Neural Networks: Tricks of the Trade
  year: 2012
  ident: PhysRevX.13.021003Cc18R1
– ident: PhysRevX.13.021003Cc35R1
  doi: 10.1088/1742-5468/ac98a7
– volume-title: Proceedings of the 12th International Conference on Artificial Intelligence and Statistics
  year: 2009
  ident: PhysRevX.13.021003Cc2R1
– ident: PhysRevX.13.021003Cc20R1
  doi: 10.1016/j.cels.2020.11.005
– ident: PhysRevX.13.021003Cc44R1
  doi: 10.1111/j.1742-4658.2008.06411.x
– volume-title: Elements of Information Theory
  year: 1999
  ident: PhysRevX.13.021003Cc30R1
– ident: PhysRevX.13.021003Cc51R1
  doi: 10.1038/s41592-022-01488-1
– ident: PhysRevX.13.021003Cc57R1
  doi: 10.1103/RevModPhys.77.579
– ident: PhysRevX.13.021003Cc11R1
  doi: 10.1109/TIP.2019.2916751
– ident: PhysRevX.13.021003Cc28R1
  doi: 10.1016/j.cpc.2020.107518
– volume-title: Proceedings of the 30th International Conference on Machine Learning
  year: 2013
  ident: PhysRevX.13.021003Cc15R1
– ident: PhysRevX.13.021003Cc29R1
  doi: 10.1088/1751-8121/ab7d00
SSID ssj0000601477
Score 2.4572747
Snippet A goal of unsupervised machine learning is to build representations of complex high-dimensional data, with simple relations to their properties. Such...
SourceID doaj
hal
proquest
crossref
SourceType Open Website
Open Access Repository
Aggregation Database
Enrichment Source
Index Database
StartPage 021003
SubjectTerms Computer Science
Constraints
Cost analysis
Datasets
Discriminators
Handwriting
Image reconstruction
Ising model
Labels
Machine learning
Neural networks
Physics
Representations
Taxonomy
Training
Two dimensional models
Unsupervised learning
SummonAdditionalLinks – databaseName: DOAJ Directory of Open Access Journals
  dbid: DOA
  link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3NS-UwEA8iCntZ1o_Fp64U8SbV5qNNevQTDyry2JV3C-kkcR9on9j6DvvX7yRtRT3oxWtIQ_hlkplf-c0MIXuZBQrgZJANilRQa9Iq8zI1xqhCQOFZzOO-vZTX12oyKW9etfoKmrCuPHAH3GElC8NzCo6rShinjJKyLBRaap5xZiPxwajnFZnq3mAM_aUcsmQyfhgElWM3nxxQfhB4ztAlq_dEsWA_-pe_QQ757lWOrub8B_nex4jJUbe3FbLg6lWyHLWa0KyR8ek0ZgyFBNz6LhlHLWufQlQ3ybTGodCNAzCYTI5n9-2_B1PXyVWUTbomCb9eZ89tEnsxN5Err5M_52e_Ty7SvjVCCoKWbapyayD3tuQukwyYLy0D5VVRCeoZUrCCh9iMGWVNIRA0hEAggNaDUN4A_0kW61ntNkhijMsBVxKCF8Jbiy4rp954cCyHylYjQgeYNPR1w0P7insd-UPG9QCtplx30I7I_ss3j13VjA9nHwf0X2aGitdxAO1A93agP7ODEdnFs3uzxsXRpQ5joawZPmxqjpO2h6PV_V1tNJOlQiKH3HPzKzayRb6FlvSdumebLLZPz-4XWYJ5O22edqKZ_gdm-eyc
  priority: 102
  providerName: Directory of Open Access Journals
Title Disentangling Representations in Restricted Boltzmann Machines without Adversaries
URI https://www.proquest.com/docview/2798096282
https://hal.science/hal-04307038
https://doaj.org/article/b76a351ce38b4ae8a8779680705032d1
Volume 13
WOSCitedRecordID wos000995916500001&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: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 2160-3308
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000601477
  issn: 2160-3308
  databaseCode: DOA
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2160-3308
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000601477
  issn: 2160-3308
  databaseCode: M~E
  dateStart: 20110101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
– providerCode: PRVPQU
  databaseName: Engineering Database
  customDbUrl:
  eissn: 2160-3308
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000601477
  issn: 2160-3308
  databaseCode: M7S
  dateStart: 20110801
  isFulltext: true
  titleUrlDefault: http://search.proquest.com
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Central
  customDbUrl:
  eissn: 2160-3308
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000601477
  issn: 2160-3308
  databaseCode: BENPR
  dateStart: 20110801
  isFulltext: true
  titleUrlDefault: https://www.proquest.com/central
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Publicly Available Content
  customDbUrl:
  eissn: 2160-3308
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000601477
  issn: 2160-3308
  databaseCode: PIMPY
  dateStart: 20110801
  isFulltext: true
  titleUrlDefault: http://search.proquest.com/publiccontent
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Science Database
  customDbUrl:
  eissn: 2160-3308
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000601477
  issn: 2160-3308
  databaseCode: M2P
  dateStart: 20110801
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/sciencejournals
  providerName: ProQuest
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1Jb9QwFLZgChIXyioG2ipC3FDaeEnsnFCni4rUGUXDouFkOV7KSCUpk3QO_Hr8PM4gOPTCJQfbsWI_-2353nsIvcuMxlpbDrBBljJsVFpnjqdKKVEwXTgS4ri_XvLZTCwWZRUdbl2EVQ48MTBq02rwkR8RXgqvbnsL4cPNzxSqRsHf1VhC4z7agUxlbIR2Jmezar71skC2Ecb5EC2T0SMAVs7tenGI6SHYO0O1rCiRQuJ-L2e-AyzyH-4cRM757v9-7BP0OCqbyfHmdDxF92zzDD0MoE_dPUfz02UIPYJI3uYqmQdQbIxFarpk2fgmKOuhvVaaTNrr_tcP1TTJNOAvbZeAD7e97ZNQ1LkLRvcL9OX87PPJRRprLKSa4bJPRW6Uzp0pqc040cSVhmjhRFEz7Ii35QoKSh5RwqiCCSX8HjJRM-M0E05p-hKNmraxr1CilM21n4kxWjBnjJd9OXbKaUtyXZt6jPCwz1LHBORQB-NaBkMko3KgjcRUbmgzRu-379xs0m_cOXoC5NuOhNTZoaFdXcl4E2XNC0VzrC31y1DWL4nzshCe9eUZJQaP0VtP_L_muDi-lNAG-dE8hxRrP2hvILyMl76Tf6j--u7uN-gRVK3fAID20Khf3dp99ECv-2W3Oohn-CC4B_xzSip48k--p_o4rb79BntQAcc
linkProvider ProQuest
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMw1V3LbtNAFL0qKQg2vFEDBSwEK-TW87A9XiDUUqpETaIoKihdDeN5tJGKXWI3CD6Kb2RmbAfBorsu2I7HI419fB_jc-8BeB0piaTUqaMN0pAiJcI8MmkohGAJlYnBvo778yidTNh8nk034FdXC-NolZ1N9IZaldKdke_iNGM23LYZwvuLb6FTjXJ_VzsJjQYWR_rHd5uyVe-GB_b9vsH48OPxh0HYqgqEkqKsDlmshIyNyoiOUiyxyRSWzLAkp8hgm70kxIU1WDAlEsoEQxGhLKfKSMqMkMSuewM2qQU768HmdDienqxPdVx3E5qmXXVORHYdkXOmV_MdRHZcftWpc7Ue0AsFWL925miY_3gD7-IO7_1vD-c-3G2D6WCvQf8D2NDFQ7jlSa2yegSzg4UvrXKVysVpMPOk37bWqqiCRWGHnGyJtFF3sF-e1z-_iqIIxp5fqqvAnVGXl3XgRasrf6jwGD5dy4aeQK8oC70FgRA6lnYlSklCjVLWt8fICCM1jmWu8j6g7r1y2TZYdzof59wnWhHhHRY4IrzBQh_eru-5aNqLXDl738FlPdO1BvcD5fKUt5aG52kiSIykJnYbQtstpWmWMGva44hghfrwyoLtrzUGeyPuxlz_N-sB2MpO2u6AxlujVvE_KHt69eWXcHtwPB7x0XBy9AzuYBsXNmSnbejVy0v9HG7KVb2oli_a7yeAL9eNyt9z4Vo3
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=Disentangling+Representations+in+Restricted+Boltzmann+Machines+without+Adversaries&rft.jtitle=Physical+review.+X&rft.au=Fernandez-De-Cossio-Diaz%2C+Jorge&rft.au=Cocco%2C+Simona&rft.au=Monasson%2C+R%C3%A9mi&rft.date=2023-04-01&rft.pub=American+Physical+Society&rft.eissn=2160-3308&rft.volume=13&rft.issue=2&rft_id=info:doi/10.1103%2FPhysRevX.13.021003&rft.externalDBID=HAS_PDF_LINK&rft.externalDocID=oai%3AHAL%3Ahal-04307038v1
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2160-3308&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2160-3308&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2160-3308&client=summon