Generative modeling via tensor train sketching

In this paper, we introduce a sketching algorithm for constructing a tensor train representation of a probability density from its samples. Our method deviates from the standard recursive SVD-based procedure for constructing a tensor train. Instead, we formulate and solve a sequence of small linear...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Applied and computational harmonic analysis Jg. 67; H. C; S. 101575
Hauptverfasser: Hur, YoonHaeng, Hoskins, Jeremy G., Lindsey, Michael, Stoudenmire, E.M., Khoo, Yuehaw
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States Elsevier Inc 01.11.2023
Elsevier
Schlagworte:
ISSN:1063-5203, 1096-603X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract In this paper, we introduce a sketching algorithm for constructing a tensor train representation of a probability density from its samples. Our method deviates from the standard recursive SVD-based procedure for constructing a tensor train. Instead, we formulate and solve a sequence of small linear systems for the individual tensor train cores. This approach can avoid the curse of dimensionality that threatens both the algorithmic and sample complexities of the recovery problem. Specifically, for Markov models under natural conditions, we prove that the tensor cores can be recovered with a sample complexity that scales logarithmically in the dimensionality. Finally, we illustrate the performance of the method with several numerical experiments.
AbstractList Not provided.
In this paper, we introduce a sketching algorithm for constructing a tensor train representation of a probability density from its samples. Our method deviates from the standard recursive SVD-based procedure for constructing a tensor train. Instead, we formulate and solve a sequence of small linear systems for the individual tensor train cores. This approach can avoid the curse of dimensionality that threatens both the algorithmic and sample complexities of the recovery problem. Specifically, for Markov models under natural conditions, we prove that the tensor cores can be recovered with a sample complexity that scales logarithmically in the dimensionality. Finally, we illustrate the performance of the method with several numerical experiments.
ArticleNumber 101575
Author Lindsey, Michael
Hur, YoonHaeng
Khoo, Yuehaw
Stoudenmire, E.M.
Hoskins, Jeremy G.
Author_xml – sequence: 1
  givenname: YoonHaeng
  orcidid: 0000-0001-6308-8896
  surname: Hur
  fullname: Hur, YoonHaeng
  email: yoonhaenghur@uchicago.edu
  organization: Department of Statistics, University of Chicago, United States of America
– sequence: 2
  givenname: Jeremy G.
  surname: Hoskins
  fullname: Hoskins, Jeremy G.
  email: jeremyhoskins@uchicago.edu
  organization: Department of Statistics, University of Chicago, United States of America
– sequence: 3
  givenname: Michael
  surname: Lindsey
  fullname: Lindsey, Michael
  email: michael.lindsey@cims.nyu.edu
  organization: Department of Mathematics, Courant Institute of Mathematical Sciences, New York University, United States of America
– sequence: 4
  givenname: E.M.
  surname: Stoudenmire
  fullname: Stoudenmire, E.M.
  email: mstoudenmire@flatironinstitute.org
  organization: Center for Computational Quantum Physics, Flatiron Institute, United States of America
– sequence: 5
  givenname: Yuehaw
  surname: Khoo
  fullname: Khoo, Yuehaw
  email: ykhoo@uchicago.edu
  organization: Department of Statistics, University of Chicago, United States of America
BackLink https://www.osti.gov/biblio/2576309$$D View this record in Osti.gov
BookMark eNp9kE9LAzEQxYNUsFa_gKfF-675s0ka8CJFq1DwouAtZLOzNrVNJAkFv70b1pOHnmaYeb_hzbtEMx88IHRDcEMwEXe7xtitaSimrAy45GdoTrAStcDsY1Z6wWpOMbtAlyntMCak5WqOmjV4iCa7I1SH0MPe-c_q6EyVwacQqxyN81X6gmy34-oKnQ9mn-D6ry7Q-9Pj2-q53ryuX1YPm9oyKnPdC0s6IYEoYrmRS0kIs1RxbISycqB9x03XgiGSd1bivgVC-0F1WIJslWnZAt1Od0PKTifrMtitDd6DzZpyKRhWo2g5iWwMKUUY9KgbXwm-uN5rgnUJR-90CUeXcPQUzojSf-h3dAcTf05D9xME4-dHB7EYA2-hd7H46oM7hf8C9vV-sw
CitedBy_id crossref_primary_10_1137_24M1654075
crossref_primary_10_1088_2632_2153_ad7571
crossref_primary_10_1016_j_jcp_2024_113110
crossref_primary_10_1103_yt8p_vr1v
crossref_primary_10_1137_24M1633765
Cites_doi 10.1137/090752286
10.1137/15M1036919
10.1111/j.1467-9868.2010.00769.x
10.1007/s10444-018-9622-8
10.1002/gamm.202100008
10.1073/pnas.0804869105
10.1017/9781316544938
10.1103/PhysRevLett.69.2863
10.1137/19M1257718
10.1137/090771806
10.1137/17M1154382
10.1007/BF01932678
10.4310/CMS.2010.v8.n1.a11
10.1007/s11222-019-09910-z
10.1103/PhysRevB.99.155131
10.1137/15M1010506
10.1016/j.laa.2009.07.024
10.1016/j.cma.2018.12.015
10.1561/2200000079
10.1137/20M1387158
ContentType Journal Article
Copyright 2023 Elsevier Inc.
Copyright_xml – notice: 2023 Elsevier Inc.
CorporateAuthor Univ. of Chicago, IL (United States)
CorporateAuthor_xml – name: Univ. of Chicago, IL (United States)
DBID AAYXX
CITATION
OTOTI
DOI 10.1016/j.acha.2023.101575
DatabaseName CrossRef
OSTI.GOV
DatabaseTitle CrossRef
DatabaseTitleList

DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Mathematics
EISSN 1096-603X
ExternalDocumentID 2576309
10_1016_j_acha_2023_101575
S1063520323000623
GroupedDBID --K
--M
.~1
0R~
1B1
1RT
1~.
1~5
23M
4.4
457
4G.
5GY
5VS
6I.
7-5
71M
8P~
9JN
AACTN
AAEDT
AAEDW
AAFTH
AAIAV
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AASFE
AAXUO
ABAOU
ABFNM
ABJNI
ABMAC
ABVKL
ABXDB
ABYKQ
ACAZW
ACDAQ
ACGFS
ACRLP
ADBBV
ADEZE
ADFGL
ADMUD
AEBSH
AEKER
AENEX
AEXQZ
AFKWA
AFTJW
AGHFR
AGUBO
AGYEJ
AHHHB
AIEXJ
AIGVJ
AIKHN
AITUG
AJBFU
AJOXV
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
ARUGR
ASPBG
AVWKF
AXJTR
AZFZN
BKOJK
BLXMC
CAG
COF
CS3
DM4
EBS
EFBJH
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-2
G-Q
GBLVA
HVGLF
HZ~
IHE
IXB
J1W
KOM
LG5
M26
M41
MCRUF
MHUIS
MO0
N9A
NCXOZ
O-L
O9-
OAUVE
OK1
OZT
P-8
P-9
P2P
PC.
Q38
R2-
RIG
ROL
RPZ
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SSW
SSZ
T5K
WUQ
XPP
ZMT
~G-
9DU
AATTM
AAXKI
AAYWO
AAYXX
ABWVN
ACLOT
ACRPL
ACVFH
ADCNI
ADNMO
ADVLN
AEIPS
AEUPX
AFJKZ
AFPUW
AGQPQ
AIGII
AIIUN
AKBMS
AKRWK
AKYEP
ANKPU
APXCP
CITATION
EFKBS
~HD
AGCQF
OTOTI
ID FETCH-LOGICAL-c327t-d6c1b67e191c5a787113c2950a69c7f2db5ab4ea175bc70d4e12df9b07e749a43
ISICitedReferencesCount 5
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001051091400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1063-5203
IngestDate Mon Aug 25 02:21:04 EDT 2025
Sat Nov 29 07:04:53 EST 2025
Tue Nov 18 22:23:38 EST 2025
Fri Feb 23 02:34:39 EST 2024
IsPeerReviewed true
IsScholarly true
Issue C
Keywords High-dimensional function approximation
Generative modeling
Tensor decompositions
Tensor train
Randomized algorithm
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c327t-d6c1b67e191c5a787113c2950a69c7f2db5ab4ea175bc70d4e12df9b07e749a43
Notes None
USDOE Office of Science (SC)
SC0022232
ORCID 0000-0001-6308-8896
0000000163088896
ParticipantIDs osti_scitechconnect_2576309
crossref_citationtrail_10_1016_j_acha_2023_101575
crossref_primary_10_1016_j_acha_2023_101575
elsevier_sciencedirect_doi_10_1016_j_acha_2023_101575
PublicationCentury 2000
PublicationDate November 2023
2023-11-00
2023-11-01
PublicationDateYYYYMMDD 2023-11-01
PublicationDate_xml – month: 11
  year: 2023
  text: November 2023
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle Applied and computational harmonic analysis
PublicationYear 2023
Publisher Elsevier Inc
Elsevier
Publisher_xml – name: Elsevier Inc
– name: Elsevier
References White (br0340) 1992; 69
Dolgov, Anaya-Izquierdo, Fox, Scheichl (br0090) 2020; 30
Kingma, Welling (br0160) 2013
Shi, Ruth, Townsend (br0260) 2021
Tabak, Vanden-Eijnden (br0300) 2010; 8
Savostyanov, Oseledets (br0250) 2011
Bigoni, Engsig-Karup, Marzouk (br0010) 2016; 38
Perez-Garcia, Verstraete, Wolf, Cirac (br0210) 2007; 7
Rezende, Mohamed (br0220) 2015
Al Daas, Ballard, Cazeaux, Hallman, Miedlar, Pasha, Reid, Saibaba (br0080) 2021
Yang, Müller, Stadtmüller (br0350) 2011; 73
Chen, Chi, Fan, Ma (br0050) 2021; 14
Gabrié, Rotskoff, Vanden-Eijnden (br0100) 2021
Al Daas, Ballard, Benner (br0070) 2022; 44
Che, Wei (br0030) 2019; 45
Novikov, Panov, Oseledets (br0180) 2021
Wedin (br0320) 1972; 12
Chen, Hoskins, Khoo, Lindsey (br0040) 2021
Süli, Mayers (br0280) 2003
Cheng, Wang, Xiang, Zhang (br0060) 2019; 99
Khoo, Lu, Ying (br0150) 2021; 19
Sun, Guo, Luo, Tropp, Udell (br0290) 2020; 2
Gorodetsky, Karaman, Marzouk (br0120) 2019; 347
Halko, Martinsson, Tropp (br0130) 2011; 53
Goodfellow, Pouget-Abadie Mehdi Mirza, Xu, Warde-Farley, Ozair, Courville, Bengio (br0110) 2014
Oseledets, Tyrtyshnikov (br0190) 2010; 432
Rokhlin, Tygert (br0230) 2008; 105
Ruthotto, Haber (br0240) 2021; 44
Oseledets (br0200) 2011; 33
Han, Wang, Fan, Wang, Zhang (br0140) 2018; 8
Nakatsukasa, Tropp (br0170) 2021
Steinlechner (br0270) 2016; 38
Wendland (br0330) 2017
Wang, Aggarwal, Aeron (br0310) 2017
Bradley, Stoudenmire, Terilla (br0020) 2020; 1
Oseledets (10.1016/j.acha.2023.101575_br0200) 2011; 33
Bigoni (10.1016/j.acha.2023.101575_br0010) 2016; 38
White (10.1016/j.acha.2023.101575_br0340) 1992; 69
Novikov (10.1016/j.acha.2023.101575_br0180) 2021
Cheng (10.1016/j.acha.2023.101575_br0060) 2019; 99
Gorodetsky (10.1016/j.acha.2023.101575_br0120) 2019; 347
Wedin (10.1016/j.acha.2023.101575_br0320) 1972; 12
Al Daas (10.1016/j.acha.2023.101575_br0080)
Han (10.1016/j.acha.2023.101575_br0140) 2018; 8
Dolgov (10.1016/j.acha.2023.101575_br0090) 2020; 30
Nakatsukasa (10.1016/j.acha.2023.101575_br0170)
Ruthotto (10.1016/j.acha.2023.101575_br0240) 2021; 44
Wang (10.1016/j.acha.2023.101575_br0310) 2017
Al Daas (10.1016/j.acha.2023.101575_br0070) 2022; 44
Süli (10.1016/j.acha.2023.101575_br0280) 2003
Chen (10.1016/j.acha.2023.101575_br0050) 2021; 14
Che (10.1016/j.acha.2023.101575_br0030) 2019; 45
Rokhlin (10.1016/j.acha.2023.101575_br0230) 2008; 105
Khoo (10.1016/j.acha.2023.101575_br0150) 2021; 19
Bradley (10.1016/j.acha.2023.101575_br0020) 2020; 1
Halko (10.1016/j.acha.2023.101575_br0130) 2011; 53
Chen (10.1016/j.acha.2023.101575_br0040)
Kingma (10.1016/j.acha.2023.101575_br0160)
Rezende (10.1016/j.acha.2023.101575_br0220) 2015
Savostyanov (10.1016/j.acha.2023.101575_br0250) 2011
Wendland (10.1016/j.acha.2023.101575_br0330) 2017
Sun (10.1016/j.acha.2023.101575_br0290) 2020; 2
Steinlechner (10.1016/j.acha.2023.101575_br0270) 2016; 38
Yang (10.1016/j.acha.2023.101575_br0350) 2011; 73
Shi (10.1016/j.acha.2023.101575_br0260)
Tabak (10.1016/j.acha.2023.101575_br0300) 2010; 8
Gabrié (10.1016/j.acha.2023.101575_br0100)
Goodfellow (10.1016/j.acha.2023.101575_br0110) 2014
Perez-Garcia (10.1016/j.acha.2023.101575_br0210) 2007; 7
Oseledets (10.1016/j.acha.2023.101575_br0190) 2010; 432
References_xml – year: 2021
  ident: br0100
  article-title: Adaptive Monte Carlo augmented with normalizing flows
– volume: 2
  start-page: 1123
  year: 2020
  end-page: 1150
  ident: br0290
  article-title: Low-rank Tucker approximation of a tensor from streaming data
  publication-title: SIAM J. Math. Data. Sci.
– volume: 1
  year: 2020
  ident: br0020
  article-title: Modeling sequences with quantum states: a look under the hood
  publication-title: Mach. Learn.: Sci. Technol.
– year: 2013
  ident: br0160
  article-title: Auto-encoding variational Bayes
– start-page: 1321
  year: 2021
  end-page: 1331
  ident: br0180
  article-title: Tensor-train density estimation
  publication-title: Uncertainty in Artificial Intelligence
– volume: 53
  start-page: 217
  year: 2011
  end-page: 288
  ident: br0130
  article-title: Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions
  publication-title: SIAM Rev.
– year: 2021
  ident: br0170
  article-title: Fast & accurate randomized algorithms for linear systems and eigenvalue problems
– volume: 38
  start-page: S461
  year: 2016
  end-page: S484
  ident: br0270
  article-title: Riemannian optimization for high-dimensional tensor completion
  publication-title: SIAM J. Sci. Comput.
– volume: 7
  start-page: 401
  year: 2007
  end-page: 430
  ident: br0210
  article-title: Matrix product state representations
  publication-title: Quantum Info. Comput.
– volume: 33
  start-page: 2295
  year: 2011
  end-page: 2317
  ident: br0200
  article-title: Tensor-train decomposition
  publication-title: SIAM J. Sci. Comput.
– volume: 8
  start-page: 217
  year: 2010
  end-page: 233
  ident: br0300
  article-title: Density estimation by dual ascent of the log-likelihood
  publication-title: Commun. Math. Sci.
– year: 2021
  ident: br0080
  article-title: Randomized algorithms for rounding in the tensor-train format
– year: 2021
  ident: br0040
  article-title: Committor functions via tensor networks
– start-page: 5697
  year: 2017
  end-page: 5705
  ident: br0310
  article-title: Efficient low rank tensor ring completion
  publication-title: International Conference on Computer Vision
– volume: 44
  start-page: C25
  year: 2022
  end-page: C53
  ident: br0070
  article-title: Parallel algorithms for tensor train arithmetic
  publication-title: SIAM J. Sci. Comput.
– year: 2003
  ident: br0280
  article-title: An Introduction to Numerical Analysis
– volume: 44
  year: 2021
  ident: br0240
  article-title: An introduction to deep generative modeling
  publication-title: GAMM-Mitteilungen
– volume: 14
  start-page: 566
  year: 2021
  end-page: 806
  ident: br0050
  article-title: Spectral methods for data science: a statistical perspective
  publication-title: Found. Trends Mach. Learn.
– volume: 432
  start-page: 70
  year: 2010
  end-page: 88
  ident: br0190
  article-title: TT-cross approximation for multidimensional arrays
  publication-title: Linear Algebra Appl.
– volume: 30
  start-page: 603
  year: 2020
  end-page: 625
  ident: br0090
  article-title: Approximation and sampling of multivariate probability distributions in the tensor train decomposition
  publication-title: Stat. Comput.
– volume: 69
  start-page: 2863
  year: 1992
  end-page: 2866
  ident: br0340
  article-title: Density matrix formulation for quantum renormalization groups
  publication-title: Phys. Rev. Lett.
– volume: 8
  year: 2018
  ident: br0140
  article-title: Unsupervised generative modeling using matrix product states
  publication-title: Phys. Rev. X.
– year: 2017
  ident: br0330
  article-title: Numerical Linear Algebra: An Introduction
  publication-title: Cambridge Texts in Applied Mathematics
– volume: 73
  start-page: 303
  year: 2011
  end-page: 324
  ident: br0350
  article-title: Functional singular component analysis
  publication-title: J. R. Stat. Soc. Ser. B: Stat. Methodol.
– start-page: 1530
  year: 2015
  end-page: 1538
  ident: br0220
  article-title: Variational inference with normalizing flows
  publication-title: International Conference on Machine Learning
– year: 2021
  ident: br0260
  article-title: Parallel algorithms for computing the tensor-train decomposition
– volume: 45
  start-page: 395
  year: 2019
  end-page: 428
  ident: br0030
  article-title: Randomized algorithms for the approximations of Tucker and the tensor train decompositions
  publication-title: Adv. Comput. Math.
– volume: 105
  start-page: 13212
  year: 2008
  end-page: 13217
  ident: br0230
  article-title: A fast randomized algorithm for overdetermined linear least-squares regression
  publication-title: Proc. Natl. Acad. Sci.
– volume: 347
  start-page: 59
  year: 2019
  end-page: 84
  ident: br0120
  article-title: A continuous analogue of the tensor-train decomposition
  publication-title: Comput. Methods Appl. Mech. Eng.
– volume: 38
  start-page: A2405
  year: 2016
  end-page: A2439
  ident: br0010
  article-title: Spectral tensor-train decomposition
  publication-title: SIAM J. Sci. Comput.
– year: 2014
  ident: br0110
  article-title: Generative adversarial nets
  publication-title: Advances in Neural Information Processing Systems
– volume: 99
  year: 2019
  ident: br0060
  article-title: Tree tensor networks for generative modeling
  publication-title: Phys. Rev. B.
– volume: 12
  start-page: 99
  year: 1972
  end-page: 111
  ident: br0320
  article-title: Perturbation bounds in connection with singular value decomposition
  publication-title: BIT Numer. Math.
– volume: 19
  start-page: 1261
  year: 2021
  end-page: 1284
  ident: br0150
  article-title: Efficient construction of tensor ring representations from sampling
  publication-title: Multiscale Model. Simul.
– start-page: 1
  year: 2011
  end-page: 8
  ident: br0250
  article-title: Fast adaptive interpolation of multi-dimensional arrays in tensor train format
  publication-title: The 2011 International Workshop on Multidimensional (nD) Systems
– volume: 8
  year: 2018
  ident: 10.1016/j.acha.2023.101575_br0140
  article-title: Unsupervised generative modeling using matrix product states
  publication-title: Phys. Rev. X.
– volume: 33
  start-page: 2295
  issue: 5
  year: 2011
  ident: 10.1016/j.acha.2023.101575_br0200
  article-title: Tensor-train decomposition
  publication-title: SIAM J. Sci. Comput.
  doi: 10.1137/090752286
– volume: 38
  start-page: A2405
  issue: 4
  year: 2016
  ident: 10.1016/j.acha.2023.101575_br0010
  article-title: Spectral tensor-train decomposition
  publication-title: SIAM J. Sci. Comput.
  doi: 10.1137/15M1036919
– start-page: 1
  year: 2011
  ident: 10.1016/j.acha.2023.101575_br0250
  article-title: Fast adaptive interpolation of multi-dimensional arrays in tensor train format
  publication-title: The 2011 International Workshop on Multidimensional (nD) Systems
– volume: 1
  issue: 3
  year: 2020
  ident: 10.1016/j.acha.2023.101575_br0020
  article-title: Modeling sequences with quantum states: a look under the hood
  publication-title: Mach. Learn.: Sci. Technol.
– year: 2014
  ident: 10.1016/j.acha.2023.101575_br0110
  article-title: Generative adversarial nets
  publication-title: Advances in Neural Information Processing Systems
– ident: 10.1016/j.acha.2023.101575_br0160
– volume: 73
  start-page: 303
  issue: 3
  year: 2011
  ident: 10.1016/j.acha.2023.101575_br0350
  article-title: Functional singular component analysis
  publication-title: J. R. Stat. Soc. Ser. B: Stat. Methodol.
  doi: 10.1111/j.1467-9868.2010.00769.x
– start-page: 5697
  year: 2017
  ident: 10.1016/j.acha.2023.101575_br0310
  article-title: Efficient low rank tensor ring completion
  publication-title: International Conference on Computer Vision
– year: 2003
  ident: 10.1016/j.acha.2023.101575_br0280
– volume: 45
  start-page: 395
  issue: 1
  year: 2019
  ident: 10.1016/j.acha.2023.101575_br0030
  article-title: Randomized algorithms for the approximations of Tucker and the tensor train decompositions
  publication-title: Adv. Comput. Math.
  doi: 10.1007/s10444-018-9622-8
– ident: 10.1016/j.acha.2023.101575_br0040
– ident: 10.1016/j.acha.2023.101575_br0080
– volume: 44
  issue: 2
  year: 2021
  ident: 10.1016/j.acha.2023.101575_br0240
  article-title: An introduction to deep generative modeling
  publication-title: GAMM-Mitteilungen
  doi: 10.1002/gamm.202100008
– ident: 10.1016/j.acha.2023.101575_br0260
– volume: 7
  start-page: 401
  issue: 5
  year: 2007
  ident: 10.1016/j.acha.2023.101575_br0210
  article-title: Matrix product state representations
  publication-title: Quantum Info. Comput.
– ident: 10.1016/j.acha.2023.101575_br0170
– volume: 105
  start-page: 13212
  issue: 36
  year: 2008
  ident: 10.1016/j.acha.2023.101575_br0230
  article-title: A fast randomized algorithm for overdetermined linear least-squares regression
  publication-title: Proc. Natl. Acad. Sci.
  doi: 10.1073/pnas.0804869105
– year: 2017
  ident: 10.1016/j.acha.2023.101575_br0330
  article-title: Numerical Linear Algebra: An Introduction
  doi: 10.1017/9781316544938
– ident: 10.1016/j.acha.2023.101575_br0100
– volume: 69
  start-page: 2863
  year: 1992
  ident: 10.1016/j.acha.2023.101575_br0340
  article-title: Density matrix formulation for quantum renormalization groups
  publication-title: Phys. Rev. Lett.
  doi: 10.1103/PhysRevLett.69.2863
– volume: 2
  start-page: 1123
  issue: 4
  year: 2020
  ident: 10.1016/j.acha.2023.101575_br0290
  article-title: Low-rank Tucker approximation of a tensor from streaming data
  publication-title: SIAM J. Math. Data. Sci.
  doi: 10.1137/19M1257718
– volume: 53
  start-page: 217
  issue: 2
  year: 2011
  ident: 10.1016/j.acha.2023.101575_br0130
  article-title: Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions
  publication-title: SIAM Rev.
  doi: 10.1137/090771806
– volume: 19
  start-page: 1261
  issue: 3
  year: 2021
  ident: 10.1016/j.acha.2023.101575_br0150
  article-title: Efficient construction of tensor ring representations from sampling
  publication-title: Multiscale Model. Simul.
  doi: 10.1137/17M1154382
– volume: 12
  start-page: 99
  issue: 1
  year: 1972
  ident: 10.1016/j.acha.2023.101575_br0320
  article-title: Perturbation bounds in connection with singular value decomposition
  publication-title: BIT Numer. Math.
  doi: 10.1007/BF01932678
– start-page: 1530
  year: 2015
  ident: 10.1016/j.acha.2023.101575_br0220
  article-title: Variational inference with normalizing flows
– volume: 8
  start-page: 217
  issue: 1
  year: 2010
  ident: 10.1016/j.acha.2023.101575_br0300
  article-title: Density estimation by dual ascent of the log-likelihood
  publication-title: Commun. Math. Sci.
  doi: 10.4310/CMS.2010.v8.n1.a11
– volume: 30
  start-page: 603
  year: 2020
  ident: 10.1016/j.acha.2023.101575_br0090
  article-title: Approximation and sampling of multivariate probability distributions in the tensor train decomposition
  publication-title: Stat. Comput.
  doi: 10.1007/s11222-019-09910-z
– volume: 99
  issue: 15
  year: 2019
  ident: 10.1016/j.acha.2023.101575_br0060
  article-title: Tree tensor networks for generative modeling
  publication-title: Phys. Rev. B.
  doi: 10.1103/PhysRevB.99.155131
– volume: 38
  start-page: S461
  issue: 5
  year: 2016
  ident: 10.1016/j.acha.2023.101575_br0270
  article-title: Riemannian optimization for high-dimensional tensor completion
  publication-title: SIAM J. Sci. Comput.
  doi: 10.1137/15M1010506
– start-page: 1321
  year: 2021
  ident: 10.1016/j.acha.2023.101575_br0180
  article-title: Tensor-train density estimation
  publication-title: Uncertainty in Artificial Intelligence
– volume: 432
  start-page: 70
  issue: 1
  year: 2010
  ident: 10.1016/j.acha.2023.101575_br0190
  article-title: TT-cross approximation for multidimensional arrays
  publication-title: Linear Algebra Appl.
  doi: 10.1016/j.laa.2009.07.024
– volume: 347
  start-page: 59
  year: 2019
  ident: 10.1016/j.acha.2023.101575_br0120
  article-title: A continuous analogue of the tensor-train decomposition
  publication-title: Comput. Methods Appl. Mech. Eng.
  doi: 10.1016/j.cma.2018.12.015
– volume: 14
  start-page: 566
  issue: 5
  year: 2021
  ident: 10.1016/j.acha.2023.101575_br0050
  article-title: Spectral methods for data science: a statistical perspective
  publication-title: Found. Trends Mach. Learn.
  doi: 10.1561/2200000079
– volume: 44
  start-page: C25
  issue: 1
  year: 2022
  ident: 10.1016/j.acha.2023.101575_br0070
  article-title: Parallel algorithms for tensor train arithmetic
  publication-title: SIAM J. Sci. Comput.
  doi: 10.1137/20M1387158
SSID ssj0011459
Score 2.4835198
Snippet In this paper, we introduce a sketching algorithm for constructing a tensor train representation of a probability density from its samples. Our method deviates...
Not provided.
SourceID osti
crossref
elsevier
SourceType Open Access Repository
Enrichment Source
Index Database
Publisher
StartPage 101575
SubjectTerms Generative modeling
High-dimensional function approximation
Mathematics
Randomized algorithm
Tensor decompositions
Tensor train
Title Generative modeling via tensor train sketching
URI https://dx.doi.org/10.1016/j.acha.2023.101575
https://www.osti.gov/biblio/2576309
Volume 67
WOSCitedRecordID wos001051091400001&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: 1096-603X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0011459
  issn: 1063-5203
  databaseCode: AIEXJ
  dateStart: 20211211
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lb9QwELZgywEOiKdoCygHbqtEsZ2s7WOFFkpFeypoOUW2Y4ttIbvabKv-_I4feVDUih64RLtRPLIynyYz45lvEPqgpLAMG5taa1haYDJLZUlYqhixTHNuc98-9v0rOznhi4XoTnRbP06ANQ2_uhLr_6pquAfKdq2z91B3LxRuwG9QOlxB7XD9J8UHImlfEeTH3LhcwOVSTl2puispdDMhpu25U1f32epYaKNHGlvd1hfbLlPo-K39qBwZOUwGNHiN_1itmkNpojhfjduex0z2kdm4lpXPWV_7s2zqeFw1LtoPM9sc0-bvmC-fZ8fZOCtBaGzPGwwpuD4Q5OZ0bGnD4I1oKsEWlGFoyl9WPCQUzgBRPx01FKHZ8PCflNk3PmV9gWFXu3ZWORmVk1EFGQ_RDmGl4BO0c_Blvjjqj5xw4Sfr9RuPHVahGPDmTm7zYiYrMMwjB-X0GXoaI4vkICDiOXpgmhfoyYhvEv4d9yS97UuUDUhJOqQkgJQkICXxSEl6pLxC3z7NTz8epnF8RqopYdu0nmmsZsxARK5LCYYZY6qJKHM5E5pZUqtSqsJIcCCVZnldGExqK1TODCuELOhrNGlWjXmDElMqVgtOdQHxN8kFBy-X50woaxSnGO8i3L2OSkduebfJX9XtithF037NOjCr3Pl02b3lKvqGweerADR3rtt3KnFrHCmydtVjsMiF2TQXe_fawz56PED9LZpsNxfmHXqkL7fLdvM-4uka2t6MjQ
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=Generative+modeling+via+tensor+train+sketching&rft.jtitle=Applied+and+computational+harmonic+analysis&rft.au=Hur%2C+YoonHaeng&rft.au=Hoskins%2C+Jeremy+G.&rft.au=Lindsey%2C+Michael&rft.au=Stoudenmire%2C+E.M.&rft.date=2023-11-01&rft.issn=1063-5203&rft.volume=67&rft.spage=101575&rft_id=info:doi/10.1016%2Fj.acha.2023.101575&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_acha_2023_101575
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1063-5203&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1063-5203&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1063-5203&client=summon