No Cover Image

Conference Paper/Proceeding/Abstract 411 views 47 downloads

Interpreting machine learning functions as physical observables

Gert Aarts Orcid Logo, Dimitrios Bachtis, Biagio Lucini Orcid Logo

Proceedings of The 38th International Symposium on Lattice Field Theory — PoS(LATTICE2021), Volume: 396

Swansea University Authors: Gert Aarts Orcid Logo, Dimitrios Bachtis, Biagio Lucini Orcid Logo

  • LATTICE2021_248.pdf

    PDF | Version of Record

    © Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0)

    Download (931.23KB)

Check full text

DOI (Published version): 10.22323/1.396.0248

Abstract

We propose to interpret machine learning functions as physical observables, opening up the possibility to apply “standard” statistical-mechanical methods to outputs from neural networks. This includes histogram reweighting and finite-size scaling, to analyse phase transitions quantitatively. In addi...

Full description

Published in: Proceedings of The 38th International Symposium on Lattice Field Theory — PoS(LATTICE2021)
ISSN: 1824-8039
Published: Trieste, Italy Sissa Medialab 2022
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa60430
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2022-07-08T19:39:18Z
last_indexed 2023-01-13T19:20:33Z
id cronfa60430
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2022-07-11T14:47:56.6363101</datestamp><bib-version>v2</bib-version><id>60430</id><entry>2022-07-08</entry><title>Interpreting machine learning functions as physical observables</title><swanseaauthors><author><sid>1ba0dad382dfe18348ec32fc65f3f3de</sid><ORCID>0000-0002-6038-3782</ORCID><firstname>Gert</firstname><surname>Aarts</surname><name>Gert Aarts</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>91a311a58d3f8badc779f0ffa6d0ca3d</sid><firstname>Dimitrios</firstname><surname>Bachtis</surname><name>Dimitrios Bachtis</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>7e6fcfe060e07a351090e2a8aba363cf</sid><ORCID>0000-0001-8974-8266</ORCID><firstname>Biagio</firstname><surname>Lucini</surname><name>Biagio Lucini</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2022-07-08</date><deptcode>SPH</deptcode><abstract>We propose to interpret machine learning functions as physical observables, opening up the possibility to apply &#x201C;standard&#x201D; statistical-mechanical methods to outputs from neural networks. This includes histogram reweighting and finite-size scaling, to analyse phase transitions quantitatively. In addition we incorporate predictive functions as conjugate variables coupled to an external field within the Hamiltonian of a system, allowing to induce order-disorder phase transitions in a novel manner. A noteworthy feature of this approach is that no knowledge of the symmetries in the Hamiltonian is required.</abstract><type>Conference Paper/Proceeding/Abstract</type><journal>Proceedings of The 38th International Symposium on Lattice Field Theory &#x2014; PoS(LATTICE2021)</journal><volume>396</volume><journalNumber/><paginationStart/><paginationEnd/><publisher>Sissa Medialab</publisher><placeOfPublication>Trieste, Italy</placeOfPublication><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic>1824-8039</issnElectronic><keywords/><publishedDay>8</publishedDay><publishedMonth>7</publishedMonth><publishedYear>2022</publishedYear><publishedDate>2022-07-08</publishedDate><doi>10.22323/1.396.0248</doi><url/><notes/><college>COLLEGE NANME</college><department>Physics</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>SPH</DepartmentCode><institution>Swansea University</institution><apcterm>Another institution paid the OA fee</apcterm><funders>ERC, STFC. Leverhulme Foundation, Royal Society, ERDF</funders><projectreference>813942, WM170010 , RF-2020-461\9, ST/T000813/1</projectreference><lastEdited>2022-07-11T14:47:56.6363101</lastEdited><Created>2022-07-08T20:18:27.9284067</Created><path><level id="1">College of Science</level><level id="2">College of Science</level></path><authors><author><firstname>Gert</firstname><surname>Aarts</surname><orcid>0000-0002-6038-3782</orcid><order>1</order></author><author><firstname>Dimitrios</firstname><surname>Bachtis</surname><order>2</order></author><author><firstname>Biagio</firstname><surname>Lucini</surname><orcid>0000-0001-8974-8266</orcid><order>3</order></author></authors><documents><document><filename>60430__24521__93aada50fb404b3fa03a3c2b494d3266.pdf</filename><originalFilename>LATTICE2021_248.pdf</originalFilename><uploaded>2022-07-08T20:38:39.9898009</uploaded><type>Output</type><contentLength>953582</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>&#xA9; Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0)</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by-nc-nd/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2022-07-11T14:47:56.6363101 v2 60430 2022-07-08 Interpreting machine learning functions as physical observables 1ba0dad382dfe18348ec32fc65f3f3de 0000-0002-6038-3782 Gert Aarts Gert Aarts true false 91a311a58d3f8badc779f0ffa6d0ca3d Dimitrios Bachtis Dimitrios Bachtis true false 7e6fcfe060e07a351090e2a8aba363cf 0000-0001-8974-8266 Biagio Lucini Biagio Lucini true false 2022-07-08 SPH We propose to interpret machine learning functions as physical observables, opening up the possibility to apply “standard” statistical-mechanical methods to outputs from neural networks. This includes histogram reweighting and finite-size scaling, to analyse phase transitions quantitatively. In addition we incorporate predictive functions as conjugate variables coupled to an external field within the Hamiltonian of a system, allowing to induce order-disorder phase transitions in a novel manner. A noteworthy feature of this approach is that no knowledge of the symmetries in the Hamiltonian is required. Conference Paper/Proceeding/Abstract Proceedings of The 38th International Symposium on Lattice Field Theory — PoS(LATTICE2021) 396 Sissa Medialab Trieste, Italy 1824-8039 8 7 2022 2022-07-08 10.22323/1.396.0248 COLLEGE NANME Physics COLLEGE CODE SPH Swansea University Another institution paid the OA fee ERC, STFC. Leverhulme Foundation, Royal Society, ERDF 813942, WM170010 , RF-2020-461\9, ST/T000813/1 2022-07-11T14:47:56.6363101 2022-07-08T20:18:27.9284067 College of Science College of Science Gert Aarts 0000-0002-6038-3782 1 Dimitrios Bachtis 2 Biagio Lucini 0000-0001-8974-8266 3 60430__24521__93aada50fb404b3fa03a3c2b494d3266.pdf LATTICE2021_248.pdf 2022-07-08T20:38:39.9898009 Output 953582 application/pdf Version of Record true © Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0) true eng https://creativecommons.org/licenses/by-nc-nd/4.0/
title Interpreting machine learning functions as physical observables
spellingShingle Interpreting machine learning functions as physical observables
Gert Aarts
Dimitrios Bachtis
Biagio Lucini
title_short Interpreting machine learning functions as physical observables
title_full Interpreting machine learning functions as physical observables
title_fullStr Interpreting machine learning functions as physical observables
title_full_unstemmed Interpreting machine learning functions as physical observables
title_sort Interpreting machine learning functions as physical observables
author_id_str_mv 1ba0dad382dfe18348ec32fc65f3f3de
91a311a58d3f8badc779f0ffa6d0ca3d
7e6fcfe060e07a351090e2a8aba363cf
author_id_fullname_str_mv 1ba0dad382dfe18348ec32fc65f3f3de_***_Gert Aarts
91a311a58d3f8badc779f0ffa6d0ca3d_***_Dimitrios Bachtis
7e6fcfe060e07a351090e2a8aba363cf_***_Biagio Lucini
author Gert Aarts
Dimitrios Bachtis
Biagio Lucini
author2 Gert Aarts
Dimitrios Bachtis
Biagio Lucini
format Conference Paper/Proceeding/Abstract
container_title Proceedings of The 38th International Symposium on Lattice Field Theory — PoS(LATTICE2021)
container_volume 396
publishDate 2022
institution Swansea University
issn 1824-8039
doi_str_mv 10.22323/1.396.0248
publisher Sissa Medialab
college_str College of Science
hierarchytype
hierarchy_top_id collegeofscience
hierarchy_top_title College of Science
hierarchy_parent_id collegeofscience
hierarchy_parent_title College of Science
department_str College of Science{{{_:::_}}}College of Science{{{_:::_}}}College of Science
document_store_str 1
active_str 0
description We propose to interpret machine learning functions as physical observables, opening up the possibility to apply “standard” statistical-mechanical methods to outputs from neural networks. This includes histogram reweighting and finite-size scaling, to analyse phase transitions quantitatively. In addition we incorporate predictive functions as conjugate variables coupled to an external field within the Hamiltonian of a system, allowing to induce order-disorder phase transitions in a novel manner. A noteworthy feature of this approach is that no knowledge of the symmetries in the Hamiltonian is required.
published_date 2022-07-08T04:18:32Z
_version_ 1763754232745295872
score 11.012678

Similar Items