A Nonstochastic Optimization Algorithm for Neural-Network Quantum States

Neural-network quantum states (NQS) employ artificial neural networks to encode many-body wave functions in a second quantization through variational Monte Carlo (VMC). They have recently been applied to accurately describe electronic wave functions of molecules and have shown the challenges in effi...

Full description

Saved in:
Bibliographic Details
Published inJournal of chemical theory and computation Vol. 19; no. 22; pp. 8156 - 8165
Main Authors Li, Xiang, Huang, Jia-Cheng, Zhang, Guang-Ze, Li, Hao-En, Cao, Chang-Su, Lv, Dingshun, Hu, Han-Shi
Format Journal Article
LanguageEnglish
Published Washington American Chemical Society 28.11.2023
Subjects
Online AccessGet full text
ISSN1549-9618
1549-9626
1549-9626
DOI10.1021/acs.jctc.3c00831

Cover

Abstract Neural-network quantum states (NQS) employ artificial neural networks to encode many-body wave functions in a second quantization through variational Monte Carlo (VMC). They have recently been applied to accurately describe electronic wave functions of molecules and have shown the challenges in efficiency compared with traditional quantum chemistry methods. Here, we introduce a general nonstochastic optimization algorithm for NQS in chemical systems, which deterministically generates a selected set of important configurations simultaneously with energy evaluation of NQS. This method bypasses the need for Markov-chain Monte Carlo within the VMC framework, thereby accelerating the entire optimization process. Furthermore, this newly developed nonstochastic optimization algorithm for NQS offers comparable or superior accuracy compared to its stochastic counterpart and ensures more stable convergence. The application of this model to test molecules exhibiting strong electron correlations provides further insight into the performance of NQS in chemical systems and opens avenues for future enhancements.
AbstractList Neural-network quantum states (NQS) employ artificial neural networks to encode many-body wave functions in a second quantization through variational Monte Carlo (VMC). They have recently been applied to accurately describe electronic wave functions of molecules and have shown the challenges in efficiency compared with traditional quantum chemistry methods. Here, we introduce a general nonstochastic optimization algorithm for NQS in chemical systems, which deterministically generates a selected set of important configurations simultaneously with energy evaluation of NQS. This method bypasses the need for Markov-chain Monte Carlo within the VMC framework, thereby accelerating the entire optimization process. Furthermore, this newly developed nonstochastic optimization algorithm for NQS offers comparable or superior accuracy compared to its stochastic counterpart and ensures more stable convergence. The application of this model to test molecules exhibiting strong electron correlations provides further insight into the performance of NQS in chemical systems and opens avenues for future enhancements.Neural-network quantum states (NQS) employ artificial neural networks to encode many-body wave functions in a second quantization through variational Monte Carlo (VMC). They have recently been applied to accurately describe electronic wave functions of molecules and have shown the challenges in efficiency compared with traditional quantum chemistry methods. Here, we introduce a general nonstochastic optimization algorithm for NQS in chemical systems, which deterministically generates a selected set of important configurations simultaneously with energy evaluation of NQS. This method bypasses the need for Markov-chain Monte Carlo within the VMC framework, thereby accelerating the entire optimization process. Furthermore, this newly developed nonstochastic optimization algorithm for NQS offers comparable or superior accuracy compared to its stochastic counterpart and ensures more stable convergence. The application of this model to test molecules exhibiting strong electron correlations provides further insight into the performance of NQS in chemical systems and opens avenues for future enhancements.
Neural-network quantum states (NQS) employ artificial neural networks to encode many-body wave functions in a second quantization through variational Monte Carlo (VMC). They have recently been applied to accurately describe electronic wave functions of molecules and have shown the challenges in efficiency compared with traditional quantum chemistry methods. Here, we introduce a general nonstochastic optimization algorithm for NQS in chemical systems, which deterministically generates a selected set of important configurations simultaneously with energy evaluation of NQS. This method bypasses the need for Markov-chain Monte Carlo within the VMC framework, thereby accelerating the entire optimization process. Furthermore, this newly developed nonstochastic optimization algorithm for NQS offers comparable or superior accuracy compared to its stochastic counterpart and ensures more stable convergence. The application of this model to test molecules exhibiting strong electron correlations provides further insight into the performance of NQS in chemical systems and opens avenues for future enhancements.
Author Li, Hao-En
Lv, Dingshun
Cao, Chang-Su
Zhang, Guang-Ze
Huang, Jia-Cheng
Li, Xiang
Hu, Han-Shi
Author_xml – sequence: 1
  givenname: Xiang
  orcidid: 0009-0004-3150-1163
  surname: Li
  fullname: Li, Xiang
  organization: Department of Chemistry and Engineering Research Center of Advanced Rare-Earth Materials of Ministry of Education, Tsinghua University, Beijing 100084, China
– sequence: 2
  givenname: Jia-Cheng
  surname: Huang
  fullname: Huang, Jia-Cheng
  organization: Department of Chemistry and Engineering Research Center of Advanced Rare-Earth Materials of Ministry of Education, Tsinghua University, Beijing 100084, China
– sequence: 3
  givenname: Guang-Ze
  surname: Zhang
  fullname: Zhang, Guang-Ze
  organization: Department of Chemistry and Engineering Research Center of Advanced Rare-Earth Materials of Ministry of Education, Tsinghua University, Beijing 100084, China
– sequence: 4
  givenname: Hao-En
  surname: Li
  fullname: Li, Hao-En
  organization: Department of Chemistry and Engineering Research Center of Advanced Rare-Earth Materials of Ministry of Education, Tsinghua University, Beijing 100084, China
– sequence: 5
  givenname: Chang-Su
  surname: Cao
  fullname: Cao, Chang-Su
  organization: Department of Chemistry and Engineering Research Center of Advanced Rare-Earth Materials of Ministry of Education, Tsinghua University, Beijing 100084, China, ByteDance Research, Zhonghang Plaza, No. 43, North Third Ring West Road, Haidian District, Beijing 100089, China
– sequence: 6
  givenname: Dingshun
  surname: Lv
  fullname: Lv, Dingshun
  organization: ByteDance Research, Zhonghang Plaza, No. 43, North Third Ring West Road, Haidian District, Beijing 100089, China
– sequence: 7
  givenname: Han-Shi
  orcidid: 0000-0001-9508-1920
  surname: Hu
  fullname: Hu, Han-Shi
  organization: Department of Chemistry and Engineering Research Center of Advanced Rare-Earth Materials of Ministry of Education, Tsinghua University, Beijing 100084, China
BookMark eNp1kE1PAjEQhhuDiYDePW7ixctiv3a3PRKiYkIgRj03s6VIcXeLbTdEf72LEA8knmYO7_Nm5hmgXuMag9A1wSOCKbkDHUYbHfWIaYwFI2eoTzIuU5nTvPe3E3GBBiFsMGaMU9ZH03Eyd02ITq8hRKuTxTba2n5DtK5JxtW78zau62TlfDI3rYcqnZu4c_4jeW6hiW2dvESIJlyi8xVUwVwd5xC9Pdy_TqbpbPH4NBnPUs0wjykYyHghMOeZZKIsBCEGG5mDKYSky6woMmCCspwC4VzmZUl0SZcl10JnogA2RLeH3q13n60JUdU2aFNV0BjXBkWFxCznGZFd9OYkunGtb7rrFJVEcs54p2GI8kNKexeCNyulbfx9P3qwlSJY7QWrTrDaC1ZHwR2IT8CttzX4r_-RH4zIgac
CitedBy_id crossref_primary_10_1360_SSC_2024_0222
crossref_primary_10_1103_PhysRevB_110_115137
crossref_primary_10_1063_5_0228731
crossref_primary_10_1063_5_0214150
Cites_doi 10.1109/TNNLS.2015.2494361
10.1103/PhysRevX.7.031059
10.1093/biomet/57.1.97
10.1103/PhysRevLett.124.097201
10.1021/acs.jctc.8b00849
10.1103/RevModPhys.32.300
10.1103/PhysRevLett.121.167204
10.1021/acs.jctc.9b01132
10.1021/acs.jctc.7b00900
10.1103/PhysRevE.74.066701
10.1103/PhysRevB.97.085104
10.1063/1.2437215
10.1063/1.1679199
10.1021/acs.jctc.9b01200
10.1021/cr2001564
10.1126/science.aag2302
10.1038/s41467-017-00705-2
10.1038/s41467-020-15724-9
10.1021/acs.jctc.5b01099
10.1103/PhysRevB.96.195145
10.1063/1.4998616
10.1103/PhysRevLett.98.110201
10.1021/acs.jctc.8b00536
10.1088/2632-2153/acdb2f
10.1038/s42256-022-00509-0
10.1063/1.4955109
10.1142/S0217979214300217
10.1103/PhysRevX.7.021021
10.1063/1.469756
10.1063/1.4948308
10.1103/PhysRevLett.124.020503
10.1021/acs.jctc.2c01216
10.1002/wcms.1340
10.1038/s41570-023-00516-8
10.1063/5.0005754
10.1021/acs.jctc.6b00407
10.1103/PhysRevX.8.011006
10.1038/s41557-020-0544-y
10.1021/acs.jctc.1c00010
10.1063/1.2746035
10.1021/acs.jctc.9b00828
10.1103/PhysRevB.96.205152
10.1140/epjd/e2014-50500-1
10.1021/acs.jctc.8b00780
10.1038/s41467-018-07520-3
10.1103/PhysRevB.104.075154
10.1103/PhysRevResearch.2.033429
10.1038/s42256-022-00461-z
10.1073/pnas.2122059119
10.1021/acs.jctc.6b01028
10.1103/RevModPhys.73.33
10.1063/5.0006074
10.1039/D2DD00093H
ContentType Journal Article
Copyright Copyright American Chemical Society Nov 28, 2023
Copyright_xml – notice: Copyright American Chemical Society Nov 28, 2023
DBID AAYXX
CITATION
7SC
7SR
7U5
8BQ
8FD
JG9
JQ2
L7M
L~C
L~D
7X8
DOI 10.1021/acs.jctc.3c00831
DatabaseName CrossRef
Computer and Information Systems Abstracts
Engineered Materials Abstracts
Solid State and Superconductivity Abstracts
METADEX
Technology Research Database
Materials Research Database
ProQuest Computer Science Collection
Advanced Technologies Database with Aerospace
Computer and Information Systems Abstracts – Academic
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
DatabaseTitle CrossRef
Materials Research Database
Engineered Materials Abstracts
Technology Research Database
Computer and Information Systems Abstracts – Academic
ProQuest Computer Science Collection
Computer and Information Systems Abstracts
Solid State and Superconductivity Abstracts
Advanced Technologies Database with Aerospace
METADEX
Computer and Information Systems Abstracts Professional
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic
Materials Research Database
DeliveryMethod fulltext_linktorsrc
Discipline Chemistry
EISSN 1549-9626
EndPage 8165
ExternalDocumentID 10_1021_acs_jctc_3c00831
GroupedDBID 4.4
53G
55A
5GY
5VS
7~N
AABXI
AAYXX
ABBLG
ABJNI
ABLBI
ABMVS
ABQRX
ABUCX
ACGFS
ACIWK
ACS
ADHLV
AEESW
AENEX
AFEFF
AHGAQ
ALMA_UNASSIGNED_HOLDINGS
AQSVZ
BAANH
CITATION
CS3
CUPRZ
D0L
DU5
EBS
ED~
F5P
GGK
GNL
IH9
J9A
JG~
P2P
RNS
ROL
UI2
VF5
VG9
W1F
7SC
7SR
7U5
8BQ
8FD
JG9
JQ2
L7M
L~C
L~D
7X8
ID FETCH-LOGICAL-c304t-aea54780445938b7811e0e96ae7892d5775a382362a14496bb1cb2db4c8c587a3
IEDL.DBID ACS
ISSN 1549-9618
1549-9626
IngestDate Fri Jul 11 07:39:00 EDT 2025
Sun Jun 29 12:53:23 EDT 2025
Thu Apr 24 22:52:09 EDT 2025
Tue Jul 01 02:03:35 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 22
Language English
License https://doi.org/10.15223/policy-029
https://doi.org/10.15223/policy-037
https://doi.org/10.15223/policy-045
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c304t-aea54780445938b7811e0e96ae7892d5775a382362a14496bb1cb2db4c8c587a3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ORCID 0009-0004-3150-1163
0000-0001-9508-1920
PQID 2919443403
PQPubID 2048741
PageCount 10
ParticipantIDs proquest_miscellaneous_2890364519
proquest_journals_2919443403
crossref_citationtrail_10_1021_acs_jctc_3c00831
crossref_primary_10_1021_acs_jctc_3c00831
ProviderPackageCode CITATION
AAYXX
PublicationCentury 2000
PublicationDate 2023-11-28
PublicationDateYYYYMMDD 2023-11-28
PublicationDate_xml – month: 11
  year: 2023
  text: 2023-11-28
  day: 28
PublicationDecade 2020
PublicationPlace Washington
PublicationPlace_xml – name: Washington
PublicationTitle Journal of chemical theory and computation
PublicationYear 2023
Publisher American Chemical Society
Publisher_xml – name: American Chemical Society
References ref9/cit9
ref45/cit45
ref3/cit3
ref27/cit27
Frank J. T. (ref56/cit56) 2021
ref16/cit16
ref52/cit52
ref23/cit23
ref8/cit8
ref31/cit31
ref2/cit2
ref34/cit34
ref37/cit37
ref20/cit20
ref48/cit48
ref17/cit17
ref10/cit10
Moreno J. R. (ref49/cit49) 2023
ref35/cit35
ref53/cit53
ref19/cit19
ref21/cit21
ref42/cit42
ref46/cit46
ref13/cit13
ref24/cit24
ref38/cit38
ref50/cit50
ref6/cit6
ref36/cit36
ref18/cit18
ref11/cit11
ref25/cit25
ref29/cit29
ref32/cit32
ref39/cit39
ref14/cit14
ref5/cit5
ref51/cit51
ref43/cit43
ref28/cit28
ref40/cit40
ref26/cit26
ref55/cit55
ref12/cit12
ref15/cit15
Martens J. (ref54/cit54) 2015; 37
ref41/cit41
ref22/cit22
ref33/cit33
ref4/cit4
ref30/cit30
ref47/cit47
ref1/cit1
ref44/cit44
ref7/cit7
References_xml – ident: ref37/cit37
  doi: 10.1109/TNNLS.2015.2494361
– ident: ref42/cit42
  doi: 10.1103/PhysRevX.7.031059
– ident: ref36/cit36
  doi: 10.1093/biomet/57.1.97
– ident: ref33/cit33
  doi: 10.1103/PhysRevLett.124.097201
– ident: ref8/cit8
  doi: 10.1021/acs.jctc.8b00849
– ident: ref45/cit45
  doi: 10.1103/RevModPhys.32.300
– ident: ref31/cit31
  doi: 10.1103/PhysRevLett.121.167204
– ident: ref17/cit17
  doi: 10.1021/acs.jctc.9b01132
– ident: ref41/cit41
  doi: 10.1021/acs.jctc.7b00900
– year: 2021
  ident: ref56/cit56
  publication-title: arXiv
– ident: ref22/cit22
  doi: 10.1103/PhysRevE.74.066701
– ident: ref34/cit34
  doi: 10.1103/PhysRevB.97.085104
– ident: ref39/cit39
  doi: 10.1063/1.2437215
– ident: ref1/cit1
  doi: 10.1063/1.1679199
– ident: ref11/cit11
  doi: 10.1021/acs.jctc.9b01200
– ident: ref27/cit27
  doi: 10.1021/cr2001564
– ident: ref15/cit15
  doi: 10.1126/science.aag2302
– volume: 37
  start-page: 2408
  year: 2015
  ident: ref54/cit54
  publication-title: Proc. 32nd Int. Conf. Machine Learning, PMLR
– ident: ref50/cit50
  doi: 10.1038/s41467-017-00705-2
– ident: ref16/cit16
  doi: 10.1038/s41467-020-15724-9
– ident: ref4/cit4
  doi: 10.1021/acs.jctc.5b01099
– ident: ref28/cit28
  doi: 10.1103/PhysRevB.96.195145
– ident: ref21/cit21
  doi: 10.1063/1.4998616
– ident: ref40/cit40
  doi: 10.1103/PhysRevLett.98.110201
– ident: ref10/cit10
  doi: 10.1021/acs.jctc.8b00536
– ident: ref25/cit25
  doi: 10.1088/2632-2153/acdb2f
– ident: ref52/cit52
  doi: 10.1038/s42256-022-00509-0
– ident: ref6/cit6
  doi: 10.1063/1.4955109
– ident: ref14/cit14
  doi: 10.1142/S0217979214300217
– ident: ref29/cit29
  doi: 10.1103/PhysRevX.7.021021
– ident: ref2/cit2
  doi: 10.1063/1.469756
– ident: ref5/cit5
  doi: 10.1063/1.4948308
– ident: ref23/cit23
  doi: 10.1103/PhysRevLett.124.020503
– ident: ref13/cit13
  doi: 10.1021/acs.jctc.2c01216
– ident: ref46/cit46
  doi: 10.1002/wcms.1340
– ident: ref20/cit20
  doi: 10.1038/s41570-023-00516-8
– ident: ref44/cit44
  doi: 10.1063/5.0005754
– ident: ref3/cit3
  doi: 10.1021/acs.jctc.6b00407
– ident: ref32/cit32
  doi: 10.1103/PhysRevX.8.011006
– ident: ref18/cit18
  doi: 10.1038/s41557-020-0544-y
– ident: ref12/cit12
  doi: 10.1021/acs.jctc.1c00010
– ident: ref38/cit38
  doi: 10.1063/1.2746035
– ident: ref9/cit9
  doi: 10.1021/acs.jctc.9b00828
– year: 2023
  ident: ref49/cit49
  publication-title: arXiv
– ident: ref30/cit30
  doi: 10.1103/PhysRevB.96.205152
– ident: ref43/cit43
  doi: 10.1140/epjd/e2014-50500-1
– ident: ref55/cit55
  doi: 10.1021/acs.jctc.8b00780
– ident: ref51/cit51
  doi: 10.1038/s41467-018-07520-3
– ident: ref35/cit35
  doi: 10.1103/PhysRevB.104.075154
– ident: ref19/cit19
  doi: 10.1103/PhysRevResearch.2.033429
– ident: ref24/cit24
  doi: 10.1038/s42256-022-00461-z
– ident: ref53/cit53
  doi: 10.1073/pnas.2122059119
– ident: ref7/cit7
  doi: 10.1021/acs.jctc.6b01028
– ident: ref26/cit26
  doi: 10.1103/RevModPhys.73.33
– ident: ref47/cit47
  doi: 10.1063/5.0006074
– ident: ref48/cit48
  doi: 10.1039/D2DD00093H
SSID ssj0033423
Score 2.4833405
Snippet Neural-network quantum states (NQS) employ artificial neural networks to encode many-body wave functions in a second quantization through variational Monte...
SourceID proquest
crossref
SourceType Aggregation Database
Enrichment Source
Index Database
StartPage 8156
SubjectTerms Algorithms
Artificial neural networks
Markov chains
Optimization
Quantum chemistry
Wave functions
Title A Nonstochastic Optimization Algorithm for Neural-Network Quantum States
URI https://www.proquest.com/docview/2919443403
https://www.proquest.com/docview/2890364519
Volume 19
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVABC
  databaseName: American Chemical Society Journals
  customDbUrl:
  eissn: 1549-9626
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0033423
  issn: 1549-9618
  databaseCode: ACS
  dateStart: 20050101
  isFulltext: true
  titleUrlDefault: https://pubs.acs.org/action/showPublications?display=journals
  providerName: American Chemical Society
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwELZQF1h4Ix4FGYmFIW3iOH6MVUVVMRQhqNQtsl2XAm2C2mTh1-PLo6iAEMxx7OjOPn_Rd_cdQldCGu2AufEiy5hHGfM9KST3QEpdaU6A_INsiwHrD-ntKBp9yuR8ZfBJ0FZm2XoxmWmFpmiLBeHWXYOAgroPddANQciukEalIDgZiIqR_GmC9RtoPQAXt0pvp2xPtCzECCGZ5LWVZ7pl3r9LNf7hg3fRdgUucafcDXtowyb7aLNb93Q7QP0OHgAgTM1UgUAzvnMRY16VYuLO7CldPGfTOXZIFoNsh5p5gzJPHN_nzgf5HJfo9BANezeP3b5X9VLwTOjTzFNWgXKXT2kkQ6GhvtT6VjJluZBkHHEeKaAEGVHuF0syrQPnxLGmRphIcBUeoUaSJvYYYSknAeMTnygmKNVcaelPAsW00My6KU5Qu7ZvbCqhceh3MYsLwpsEsbNQDBaKKwudoOvVG2-lyMYvY5u1y-LquC1jIgNJaUh9t_jl6rGzLbAfKrFp7sYIWXCugTz9x3JnaAsazEP1IRFN1MgWuT13MCTTF8UG_AAMtdfK
linkProvider American Chemical Society
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=A+Nonstochastic+Optimization+Algorithm+for+Neural-Network+Quantum+States&rft.jtitle=Journal+of+chemical+theory+and+computation&rft.au=Li%2C+Xiang&rft.au=Huang%2C+Jia-Cheng&rft.au=Zhang%2C+Guang-Ze&rft.au=Li%2C+Hao-En&rft.date=2023-11-28&rft.pub=American+Chemical+Society&rft.issn=1549-9618&rft.eissn=1549-9626&rft.volume=19&rft.issue=22&rft.spage=8156&rft_id=info:doi/10.1021%2Facs.jctc.3c00831&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1549-9618&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1549-9618&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1549-9618&client=summon