Neural Network Renormalization Group
We present a variational renormalization group (RG) approach based on a reversible generative model with hierarchical architecture. The model performs hierarchical change-of-variables transformations from the physical space to a latent space with reduced mutual information. Conversely, the neural ne...
Uloženo v:
| Vydáno v: | Physical review letters Ročník 121; číslo 26; s. 260601 |
|---|---|
| Hlavní autoři: | , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
United States
American Physical Society
28.12.2018
|
| Témata: | |
| ISSN: | 0031-9007, 1079-7114, 1079-7114 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | We present a variational renormalization group (RG) approach based on a reversible generative model with hierarchical architecture. The model performs hierarchical change-of-variables transformations from the physical space to a latent space with reduced mutual information. Conversely, the neural network directly maps independent Gaussian noises to physical configurations following the inverse RG flow. The model has an exact and tractable likelihood, which allows unbiased training and direct access to the renormalized energy function of the latent variables. To train the model, we employ probability density distillation for the bare energy function of the physical problem, in which the training loss provides a variational upper bound of the physical free energy. We demonstrate practical usage of the approach by identifying mutually independent collective variables of the Ising model and performing accelerated hybrid Monte Carlo sampling in the latent space. Lastly, we comment on the connection of the present approach to the wavelet formulation of RG and the modern pursuit of information preserving RG. |
|---|---|
| AbstractList | We present a variational renormalization group (RG) approach based on a reversible generative model with hierarchical architecture. The model performs hierarchical change-of-variables transformations from the physical space to a latent space with reduced mutual information. Conversely, the neural network directly maps independent Gaussian noises to physical configurations following the inverse RG flow. The model has an exact and tractable likelihood, which allows unbiased training and direct access to the renormalized energy function of the latent variables. To train the model, we employ probability density distillation for the bare energy function of the physical problem, in which the training loss provides a variational upper bound of the physical free energy. We demonstrate practical usage of the approach by identifying mutually independent collective variables of the Ising model and performing accelerated hybrid Monte Carlo sampling in the latent space. Lastly, we comment on the connection of the present approach to the wavelet formulation of RG and the modern pursuit of information preserving RG.We present a variational renormalization group (RG) approach based on a reversible generative model with hierarchical architecture. The model performs hierarchical change-of-variables transformations from the physical space to a latent space with reduced mutual information. Conversely, the neural network directly maps independent Gaussian noises to physical configurations following the inverse RG flow. The model has an exact and tractable likelihood, which allows unbiased training and direct access to the renormalized energy function of the latent variables. To train the model, we employ probability density distillation for the bare energy function of the physical problem, in which the training loss provides a variational upper bound of the physical free energy. We demonstrate practical usage of the approach by identifying mutually independent collective variables of the Ising model and performing accelerated hybrid Monte Carlo sampling in the latent space. Lastly, we comment on the connection of the present approach to the wavelet formulation of RG and the modern pursuit of information preserving RG. We present a variational renormalization group (RG) approach based on a reversible generative model with hierarchical architecture. The model performs hierarchical change-of-variables transformations from the physical space to a latent space with reduced mutual information. Conversely, the neural network directly maps independent Gaussian noises to physical configurations following the inverse RG flow. The model has an exact and tractable likelihood, which allows unbiased training and direct access to the renormalized energy function of the latent variables. To train the model, we employ probability density distillation for the bare energy function of the physical problem, in which the training loss provides a variational upper bound of the physical free energy. We demonstrate practical usage of the approach by identifying mutually independent collective variables of the Ising model and performing accelerated hybrid Monte Carlo sampling in the latent space. Lastly, we comment on the connection of the present approach to the wavelet formulation of RG and the modern pursuit of information preserving RG. We present a variational renormalization group (RG) approach based on a reversible generative model with hierarchical architecture. The model performs hierarchical change-of-variables transformations from the physical space to a latent space with reduced mutual information. Conversely, the neural network directly maps independent Gaussian noises to physical configurations following the inverse RG flow. The model has an exact and tractable likelihood, which allows unbiased training and direct access to the renormalized energy function of the latent variables. To train the model, we employ probability density distillation for the bare energy function of the physical problem, in which the training loss provides a variational upper bound of the physical free energy. We demonstrate practical usage of the approach by identifying mutually independent collective variables of the Ising model and performing accelerated hybrid Monte Carlo sampling in the latent space. Lastly, we comment on the connection of the present approach to the wavelet formulation of RG and the modern pursuit of information preserving RG. |
| ArticleNumber | 260601 |
| Author | Li, Shuo-Hui Wang, Lei |
| Author_xml | – sequence: 1 givenname: Shuo-Hui surname: Li fullname: Li, Shuo-Hui – sequence: 2 givenname: Lei surname: Wang fullname: Wang, Lei |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/30636161$$D View this record in MEDLINE/PubMed |
| BookMark | eNqFkE1Lw0AQhhep2A_9C6WgBy-pM7vpbgJepGgVSpWi52Wz2WBqkq27iVJ_valtQbx4msvzzsz79EmnspUhZIgwRgR29fS68UvzMTd1PUaKY8qBAx6RHoKIA4EYdkgPgGEQA4gu6Xu_AgCkPDohXQacceTYIxcL0zhVjBam_rTubbQ0lXWlKvIvVee2Gs2cbdan5DhThTdn-zkgL3e3z9P7YP44e5jezAPNWFwHqeFMxWGmmYBQJYInLFE6yZTmXMCEU0aFzqjChLEUTYrUUBNiqDREQscJG5DL3d61s--N8bUsc69NUajK2MZLiiJmE4g4tuj5H3RlG1e1320pFoUYC9ZSwz3VJKVJ5drlpXIbeejfAtc7QDvrvTOZ1Hn907x2Ki8kgtzqlr90y1a33Olu4_xP_HDhn-A39hGGMQ |
| CitedBy_id | crossref_primary_10_1103_PhysRevResearch_6_043322 crossref_primary_10_1140_epja_s10050_023_01154_w crossref_primary_10_1103_PhysRevX_10_021020 crossref_primary_10_1134_S1063779624702162 crossref_primary_10_1016_j_physd_2024_134505 crossref_primary_10_1103_PhysRevResearch_6_033041 crossref_primary_10_3390_e22050587 crossref_primary_10_3390_sym14030486 crossref_primary_10_1088_2632_2153_ad0101 crossref_primary_10_7566_JPSJ_93_064002 crossref_primary_10_1088_0256_307X_39_5_050701 crossref_primary_10_1080_23746149_2020_1797528 crossref_primary_10_1016_j_ppnp_2023_104084 crossref_primary_10_1103_PhysRevX_13_041038 crossref_primary_10_1088_1361_648X_abb895 crossref_primary_10_1016_j_jcp_2025_113806 crossref_primary_10_1088_1742_5468_ad5c5c crossref_primary_10_1038_s41524_022_00736_4 crossref_primary_10_1088_0256_307X_40_2_020501 crossref_primary_10_1007_JHEP07_2022_015 crossref_primary_10_1088_0256_307X_40_12_120201 crossref_primary_10_1002_esp_5984 crossref_primary_10_1103_PRXQuantum_2_040201 crossref_primary_10_1007_s10255_025_0001_1 crossref_primary_10_1103_PhysRevResearch_2_023369 crossref_primary_10_1088_1367_2630_acef4e crossref_primary_10_1002_ett_4865 crossref_primary_10_1088_1751_8121_ad72ba crossref_primary_10_1103_rbsg_r7hx crossref_primary_10_1038_s41598_021_85683_8 crossref_primary_10_1007_s10489_019_01546_w crossref_primary_10_1007_s11633_022_1340_5 crossref_primary_10_1088_2632_2153_acb488 crossref_primary_10_3389_frai_2020_00030 crossref_primary_10_1103_PhysRevResearch_3_L042024 crossref_primary_10_1103_PhysRevResearch_4_L042005 crossref_primary_10_1109_TGRS_2023_3304297 crossref_primary_10_7566_JPSJ_91_062001 crossref_primary_10_1088_2632_2153_ac48a2 crossref_primary_10_1088_2632_2153_ac4f69 crossref_primary_10_1103_PhysRevX_10_011037 crossref_primary_10_1088_2632_2153_acbe91 crossref_primary_10_1103_PhysRevB_105_205139 crossref_primary_10_3390_e25010026 crossref_primary_10_1016_j_commatsci_2022_111634 crossref_primary_10_1103_8wx7_kyx8 crossref_primary_10_1063_5_0018903 crossref_primary_10_7566_JPSJ_88_054002 crossref_primary_10_3390_e23010123 crossref_primary_10_1103_PhysRevE_101_023304 crossref_primary_10_34133_icomputing_0123 crossref_primary_10_1016_j_physa_2022_128276 crossref_primary_10_1103_PhysRevResearch_2_023266 crossref_primary_10_1103_PhysRevResearch_2_023300 crossref_primary_10_1103_PhysRevX_11_031059 crossref_primary_10_1038_s42005_025_02261_4 crossref_primary_10_1088_2632_2153_ac8393 crossref_primary_10_1145_3654662 crossref_primary_10_7566_JPSJ_89_022001 crossref_primary_10_1137_24M1666628 crossref_primary_10_1103_PhysRevResearch_3_013134 crossref_primary_10_1016_j_neunet_2025_108126 crossref_primary_10_1038_s41598_021_88605_w |
| Cites_doi | 10.1007/s10955-017-1770-6 10.1063/1.1543581 10.1103/RevModPhys.47.773 10.1103/PhysRevLett.116.140403 10.1016/0370-2693(87)91197-X 10.1103/PhysRevLett.118.110504 10.1063/1.1543582 10.1103/PhysRevB.80.155131 10.1103/PhysRevB.95.041101 10.1103/PhysRevLett.101.110501 10.1103/PhysRevLett.103.160601 10.1103/RevModPhys.86.647 10.1103/PhysRevLett.105.010502 10.1103/PhysRevLett.99.120601 10.1103/PhysRevB.86.045139 10.1103/PhysRevE.97.053304 10.1103/PhysRevB.4.3174 10.1103/PhysRevLett.115.200401 10.1103/PhysRevB.81.174411 10.1103/PhysRevE.69.066138 10.1103/PhysRev.65.117 10.1103/PhysRevB.97.045111 10.1038/s41567-018-0081-4 10.1103/PhysRevLett.91.147902 10.1103/PhysRevD.86.065007 10.1007/s10955-017-1836-5 10.1103/PhysRev.95.1300 10.1103/PhysRevLett.42.859 10.1103/PhysRevB.97.045153 10.1103/PhysRev.76.1232 10.1103/PhysRevA.97.052314 10.1073/pnas.1618455114 10.1103/PhysRevA.74.022320 10.1103/PhysRevB.4.3184 10.1103/PhysRevLett.100.070502 10.1103/PhysRevB.95.035105 10.1063/1.1699114 10.1103/PhysRevLett.115.180405 10.1103/PhysRevB.78.205116 10.1103/PhysRevLett.118.250602 10.1002/wcms.31 |
| ContentType | Journal Article |
| Copyright | Copyright American Physical Society Dec 28, 2018 |
| Copyright_xml | – notice: Copyright American Physical Society Dec 28, 2018 |
| DBID | AAYXX CITATION NPM 7U5 8FD H8D L7M 7X8 |
| DOI | 10.1103/PhysRevLett.121.260601 |
| DatabaseName | CrossRef PubMed Solid State and Superconductivity Abstracts Technology Research Database Aerospace Database Advanced Technologies Database with Aerospace MEDLINE - Academic |
| DatabaseTitle | CrossRef PubMed Aerospace Database Solid State and Superconductivity Abstracts Technology Research Database Advanced Technologies Database with Aerospace MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic PubMed Aerospace Database |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Physics |
| EISSN | 1079-7114 |
| ExternalDocumentID | 30636161 10_1103_PhysRevLett_121_260601 |
| Genre | Journal Article |
| GroupedDBID | --- -DZ -~X 123 186 2-P 29O 3MX 3O- 41~ 5VS 6TJ 85S 8NH 8WZ 9M8 A6W AAYJJ AAYXX ABSSX ABUFD ACBEA ACGFO ACKIV ACNCT ADXHL AECSF AENEX AEQTI AETEA AFFNX AFGMR AGDNE AJQPL ALMA_UNASSIGNED_HOLDINGS APKKM AUAIK CITATION CS3 D0L DU5 EBS EJD ER. F5P H~9 MVM N9A NEJ NHB NPBMV OHT OK1 P0- P2P RNS ROL S7W SJN T9H TN5 UBC UBE VOH WH7 XOL XSW YNT YYP ZCG ZPR ZY4 ~02 NPM UCJ VQA 7U5 8FD H8D L7M 7X8 |
| ID | FETCH-LOGICAL-c339t-de63a94fc3704ab76b3bacbfac6670562327cf2a1b33d1ed12e2e414ac087c9b3 |
| IEDL.DBID | 3MX |
| ISICitedReferencesCount | 113 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000454432900001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0031-9007 1079-7114 |
| IngestDate | Fri Jul 11 09:23:42 EDT 2025 Mon Jun 30 02:35:25 EDT 2025 Thu Jan 02 23:01:52 EST 2025 Tue Nov 18 20:57:49 EST 2025 Sat Nov 29 05:55:17 EST 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 26 |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c339t-de63a94fc3704ab76b3bacbfac6670562327cf2a1b33d1ed12e2e414ac087c9b3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23 |
| PMID | 30636161 |
| PQID | 2173841973 |
| PQPubID | 2048222 |
| ParticipantIDs | proquest_miscellaneous_2179350861 proquest_journals_2173841973 pubmed_primary_30636161 crossref_citationtrail_10_1103_PhysRevLett_121_260601 crossref_primary_10_1103_PhysRevLett_121_260601 |
| PublicationCentury | 2000 |
| PublicationDate | 2018-Dec-28 |
| PublicationDateYYYYMMDD | 2018-12-28 |
| PublicationDate_xml | – month: 12 year: 2018 text: 2018-Dec-28 day: 28 |
| PublicationDecade | 2010 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States – name: College Park |
| PublicationTitle | Physical review letters |
| PublicationTitleAlternate | Phys Rev Lett |
| PublicationYear | 2018 |
| Publisher | American Physical Society |
| Publisher_xml | – name: American Physical Society |
| References | J. S. Liu (PhysRevLett.121.260601Cc62R1) 2001 PhysRevLett.121.260601Cc10R1 PhysRevLett.121.260601Cc56R1 PhysRevLett.121.260601Cc79R1 M. E. Fisher (PhysRevLett.121.260601Cc68R1) 1983 PhysRevLett.121.260601Cc14R1 PhysRevLett.121.260601Cc12R1 PhysRevLett.121.260601Cc54R1 PhysRevLett.121.260601Cc75R1 A. Paszke (PhysRevLett.121.260601Cc52R1) 2017 R. M. Neal (PhysRevLett.121.260601Cc45R1) 2011 PhysRevLett.121.260601Cc71R1 I. Goodfellow (PhysRevLett.121.260601Cc50R1) 2016 B. Guy (PhysRevLett.121.260601Cc73R1) 1999 R. P. Feynman (PhysRevLett.121.260601Cc63R1) 1972 PhysRevLett.121.260601Cc18R1 PhysRevLett.121.260601Cc16R1 PhysRevLett.121.260601Cc42R1 PhysRevLett.121.260601Cc67R1 PhysRevLett.121.260601Cc25R1 PhysRevLett.121.260601Cc9R1 PhysRevLett.121.260601Cc23R1 PhysRevLett.121.260601Cc7R1 PhysRevLett.121.260601Cc5R1 PhysRevLett.121.260601Cc80R1 PhysRevLett.121.260601Cc3R1 PhysRevLett.121.260601Cc82R1 PhysRevLett.121.260601Cc1R1 D. J. C. MacKay (PhysRevLett.121.260601Cc61R1) 2005 PhysRevLett.121.260601Cc27R1 D. A. Moore (PhysRevLett.121.260601Cc48R1) 2016 PhysRevLett.121.260601Cc55R1 PhysRevLett.121.260601Cc57R1 PhysRevLett.121.260601Cc13R1 PhysRevLett.121.260601Cc11R1 PhysRevLett.121.260601Cc76R1 PhysRevLett.121.260601Cc72R1 C. M. Bishop (PhysRevLett.121.260601Cc64R1) 2006 PhysRevLett.121.260601Cc17R1 PhysRevLett.121.260601Cc15R1 PhysRevLett.121.260601Cc59R1 PhysRevLett.121.260601Cc43R1 PhysRevLett.121.260601Cc66R1 PhysRevLett.121.260601Cc8R1 PhysRevLett.121.260601Cc24R1 PhysRevLett.121.260601Cc6R1 PhysRevLett.121.260601Cc22R1 PhysRevLett.121.260601Cc4R1 PhysRevLett.121.260601Cc81R1 M. D. Zeiler (PhysRevLett.121.260601Cc19R1) 2014 PhysRevLett.121.260601Cc2R1 PhysRevLett.121.260601Cc60R1 Y. Zhang (PhysRevLett.121.260601Cc69R1) 2012; 25 |
| References_xml | – ident: PhysRevLett.121.260601Cc22R1 doi: 10.1007/s10955-017-1770-6 – volume-title: Proceedings of the European Conference on Computer Vision year: 2014 ident: PhysRevLett.121.260601Cc19R1 – ident: PhysRevLett.121.260601Cc66R1 doi: 10.1063/1.1543581 – volume-title: Wavelets and Renormalization year: 1999 ident: PhysRevLett.121.260601Cc73R1 – ident: PhysRevLett.121.260601Cc4R1 doi: 10.1103/RevModPhys.47.773 – ident: PhysRevLett.121.260601Cc75R1 doi: 10.1103/PhysRevLett.116.140403 – ident: PhysRevLett.121.260601Cc43R1 doi: 10.1016/0370-2693(87)91197-X – ident: PhysRevLett.121.260601Cc16R1 doi: 10.1103/PhysRevLett.118.110504 – volume-title: Critical Phenomena year: 1983 ident: PhysRevLett.121.260601Cc68R1 – volume-title: NIPS 2017 Workshop Autodiff year: 2017 ident: PhysRevLett.121.260601Cc52R1 – volume-title: Pattern Recognition and Machine Learning year: 2006 ident: PhysRevLett.121.260601Cc64R1 – ident: PhysRevLett.121.260601Cc67R1 doi: 10.1063/1.1543582 – ident: PhysRevLett.121.260601Cc9R1 doi: 10.1103/PhysRevB.80.155131 – volume-title: Monte Carlo Strategies in Scientific Computing year: 2001 ident: PhysRevLett.121.260601Cc62R1 – ident: PhysRevLett.121.260601Cc60R1 doi: 10.1103/PhysRevB.95.041101 – ident: PhysRevLett.121.260601Cc6R1 doi: 10.1103/PhysRevLett.101.110501 – ident: PhysRevLett.121.260601Cc10R1 doi: 10.1103/PhysRevLett.103.160601 – ident: PhysRevLett.121.260601Cc13R1 doi: 10.1103/RevModPhys.86.647 – ident: PhysRevLett.121.260601Cc56R1 doi: 10.1103/PhysRevLett.105.010502 – ident: PhysRevLett.121.260601Cc7R1 doi: 10.1103/PhysRevLett.99.120601 – ident: PhysRevLett.121.260601Cc12R1 doi: 10.1103/PhysRevB.86.045139 – ident: PhysRevLett.121.260601Cc23R1 doi: 10.1103/PhysRevE.97.053304 – ident: PhysRevLett.121.260601Cc2R1 doi: 10.1103/PhysRevB.4.3174 – ident: PhysRevLett.121.260601Cc15R1 doi: 10.1103/PhysRevLett.115.200401 – ident: PhysRevLett.121.260601Cc11R1 doi: 10.1103/PhysRevB.81.174411 – ident: PhysRevLett.121.260601Cc79R1 doi: 10.1103/PhysRevE.69.066138 – ident: PhysRevLett.121.260601Cc71R1 doi: 10.1103/PhysRev.65.117 – ident: PhysRevLett.121.260601Cc18R1 doi: 10.1103/PhysRevB.97.045111 – ident: PhysRevLett.121.260601Cc24R1 doi: 10.1038/s41567-018-0081-4 – ident: PhysRevLett.121.260601Cc54R1 doi: 10.1103/PhysRevLett.91.147902 – volume-title: Handbook of Markov Chain Monte Carlo year: 2011 ident: PhysRevLett.121.260601Cc45R1 – ident: PhysRevLett.121.260601Cc82R1 doi: 10.1103/PhysRevD.86.065007 – ident: PhysRevLett.121.260601Cc27R1 doi: 10.1007/s10955-017-1836-5 – ident: PhysRevLett.121.260601Cc1R1 doi: 10.1103/PhysRev.95.1300 – ident: PhysRevLett.121.260601Cc5R1 doi: 10.1103/PhysRevLett.42.859 – ident: PhysRevLett.121.260601Cc25R1 doi: 10.1103/PhysRevB.97.045153 – volume: 25 start-page: 3194 issn: 1049-5258 year: 2012 ident: PhysRevLett.121.260601Cc69R1 publication-title: Adv. Neural Inf. Process. Syst. – ident: PhysRevLett.121.260601Cc72R1 doi: 10.1103/PhysRev.76.1232 – volume-title: Deep Learning year: 2016 ident: PhysRevLett.121.260601Cc50R1 – ident: PhysRevLett.121.260601Cc76R1 doi: 10.1103/PhysRevA.97.052314 – ident: PhysRevLett.121.260601Cc81R1 doi: 10.1073/pnas.1618455114 – ident: PhysRevLett.121.260601Cc55R1 doi: 10.1103/PhysRevA.74.022320 – ident: PhysRevLett.121.260601Cc3R1 doi: 10.1103/PhysRevB.4.3184 – ident: PhysRevLett.121.260601Cc57R1 doi: 10.1103/PhysRevLett.100.070502 – volume-title: Statistical Mechanics: A Set of Lectures year: 1972 ident: PhysRevLett.121.260601Cc63R1 – ident: PhysRevLett.121.260601Cc59R1 doi: 10.1103/PhysRevB.95.035105 – ident: PhysRevLett.121.260601Cc42R1 doi: 10.1063/1.1699114 – ident: PhysRevLett.121.260601Cc14R1 doi: 10.1103/PhysRevLett.115.180405 – volume-title: Information Theory, Inference, and Learning Algorithms year: 2005 ident: PhysRevLett.121.260601Cc61R1 – ident: PhysRevLett.121.260601Cc8R1 doi: 10.1103/PhysRevB.78.205116 – volume-title: NIPS Workshop on Advances in Approximate Bayesian Inference year: 2016 ident: PhysRevLett.121.260601Cc48R1 – ident: PhysRevLett.121.260601Cc17R1 doi: 10.1103/PhysRevLett.118.250602 – ident: PhysRevLett.121.260601Cc80R1 doi: 10.1002/wcms.31 |
| SSID | ssj0001268 |
| Score | 2.6531312 |
| Snippet | We present a variational renormalization group (RG) approach based on a reversible generative model with hierarchical architecture. The model performs... |
| SourceID | proquest pubmed crossref |
| SourceType | Aggregation Database Index Database Enrichment Source |
| StartPage | 260601 |
| SubjectTerms | Computer simulation Distillation Free energy Independent variables Ising model Mathematical models Neural networks Training Upper bounds Wavelet analysis |
| Title | Neural Network Renormalization Group |
| URI | https://www.ncbi.nlm.nih.gov/pubmed/30636161 https://www.proquest.com/docview/2173841973 https://www.proquest.com/docview/2179350861 |
| Volume | 121 |
| WOSCitedRecordID | wos000454432900001&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: PRVABR databaseName: American Physical Society Journals customDbUrl: eissn: 1079-7114 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0001268 issn: 0031-9007 databaseCode: 3MX dateStart: 20020101 isFulltext: true titleUrlDefault: https://journals.aps.org/ providerName: American Physical Society – providerCode: PRVIAO databaseName: SCOAP3 Journals customDbUrl: eissn: 1079-7114 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0001268 issn: 0031-9007 databaseCode: ER. dateStart: 20180101 isFulltext: true titleUrlDefault: https://scoap3.org/ providerName: SCOAP3 (Sponsoring Consortium for Open Access Publishing in Particle Physics) |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dS8MwEA9jKPji98d0SoW91jW9Nh-PIg5fHDIU-laS9ArC6GRff79J2hUFh-ylL2na43KX-11yH4QMrIkrMdYYQmElOCmptjrH01DKVChZUANR4ZtN8PFYZJl865Do7xt8GsHQRUJOcO2yW1wphAcLwFmdsCUS17IAXrN266Uxq7decHEHEW9Sgrd_5rc12gIxvakZHe1O5DE5bGBl8FjLwQnpYHVK9n14p1mckYGrwWHHx3XQdzDBymHVaZOEGfgTqHPyMXp-f3oJm_4IoQGQy7BABkompQEeJUpzpkEro0tlGOMe2MTclLGiGqCgWNAYY0xookwkuJEaLki3mlV4RYJCppyqApQr8GaQaUxlgoBCUVSa0R5JN3zKTVM83PWwmObeiYgg_8GB3HIgrznQI8N23lddPuPfGf3NMuSNOi1y6zeBSKjk0CP37bBVBHe7oSqcrfw7EizcdMRe1svX_tL6RcAstr3emZwbcmAhknABLLHok-5yvsJbsmfWy8_F_M5Ln33yTHwDWDvXaw |
| linkProvider | American Physical 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=Neural+Network+Renormalization+Group&rft.jtitle=Physical+review+letters&rft.au=Li%2C+Shuo-Hui&rft.au=Wang%2C+Lei&rft.date=2018-12-28&rft.issn=1079-7114&rft.eissn=1079-7114&rft.volume=121&rft.issue=26&rft.spage=260601&rft_id=info:doi/10.1103%2FPhysRevLett.121.260601&rft.externalDBID=NO_FULL_TEXT |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0031-9007&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0031-9007&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0031-9007&client=summon |