Fast and accurate modeling of molecular atomization energies with machine learning

We introduce a machine learning model to predict atomization energies of a diverse set of organic molecules, based on nuclear charges and atomic positions only. The problem of solving the molecular Schrödinger equation is mapped onto a nonlinear statistical regression problem of reduced complexity....

Full description

Saved in:
Bibliographic Details
Published in:Physical review letters 2012-01, Vol.108 (5), p.058301-058301, Article 058301
Main Authors: Rupp, Matthias, Tkatchenko, Alexandre, Müller, Klaus-Robert, von Lilienfeld, O Anatole
Format: Article
Language:English
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c438t-485466b7f94157f54d81304f62a7b86c9f8729efd7947cd5534d46fccdcc78453
cites cdi_FETCH-LOGICAL-c438t-485466b7f94157f54d81304f62a7b86c9f8729efd7947cd5534d46fccdcc78453
container_end_page 058301
container_issue 5
container_start_page 058301
container_title Physical review letters
container_volume 108
creator Rupp, Matthias
Tkatchenko, Alexandre
Müller, Klaus-Robert
von Lilienfeld, O Anatole
description We introduce a machine learning model to predict atomization energies of a diverse set of organic molecules, based on nuclear charges and atomic positions only. The problem of solving the molecular Schrödinger equation is mapped onto a nonlinear statistical regression problem of reduced complexity. Regression models are trained on and compared to atomization energies computed with hybrid density-functional theory. Cross validation over more than seven thousand organic molecules yields a mean absolute error of ∼10  kcal/mol. Applicability is demonstrated for the prediction of molecular atomization potential energy curves.
doi_str_mv 10.1103/PhysRevLett.108.058301
format article
fullrecord <record><control><sourceid>proquest_osti_</sourceid><recordid>TN_cdi_osti_scitechconnect_1099077</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>927688924</sourcerecordid><originalsourceid>FETCH-LOGICAL-c438t-485466b7f94157f54d81304f62a7b86c9f8729efd7947cd5534d46fccdcc78453</originalsourceid><addsrcrecordid>eNpNkF1rFTEQhoNY7Gn1L5TgjVd7nOxm83EpxVbhgKXo9ZIzO-mJ7CY1yVrqr-_KqcWrGYbnfQcexi4EbIWA7uPN4bHc0u8d1boVYLbQmw7EK7YRoG2jhZCv2QagE40F0KfsrJSfACBaZd6w07aVAFbpDbu9cqVyF0fuEJfsKvE5jTSFeMeTX_eJcJlc5q6mOfxxNaTIKVK-C1T4Q6gHPjs8hEh8IpfjmnvLTrybCr17nufsx9Xn75dfmt2366-Xn3YNys7URppeKrXX3krRa9_L0YgOpFet03uj0HqjW0t-1FZqHPu-k6NUHnFE1Eb23Tl7f-xNpYahYKiEB0wxEtZBgLWg9Qp9OEL3Of1aqNRhDgVpmlyktJTBtloZY1u5kupIYk6lZPLDfQ6zy49r1_DX-fCf8_VmhqPzNXjx_GLZzzS-xP5J7p4AgiKAMw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>927688924</pqid></control><display><type>article</type><title>Fast and accurate modeling of molecular atomization energies with machine learning</title><source>American Physical Society:Jisc Collections:APS Read and Publish 2023-2025 (reading list)</source><creator>Rupp, Matthias ; Tkatchenko, Alexandre ; Müller, Klaus-Robert ; von Lilienfeld, O Anatole</creator><creatorcontrib>Rupp, Matthias ; Tkatchenko, Alexandre ; Müller, Klaus-Robert ; von Lilienfeld, O Anatole</creatorcontrib><description>We introduce a machine learning model to predict atomization energies of a diverse set of organic molecules, based on nuclear charges and atomic positions only. The problem of solving the molecular Schrödinger equation is mapped onto a nonlinear statistical regression problem of reduced complexity. Regression models are trained on and compared to atomization energies computed with hybrid density-functional theory. Cross validation over more than seven thousand organic molecules yields a mean absolute error of ∼10  kcal/mol. Applicability is demonstrated for the prediction of molecular atomization potential energy curves.</description><identifier>ISSN: 0031-9007</identifier><identifier>EISSN: 1079-7114</identifier><identifier>DOI: 10.1103/PhysRevLett.108.058301</identifier><identifier>PMID: 22400967</identifier><language>eng</language><publisher>United States: American Physical Society</publisher><ispartof>Physical review letters, 2012-01, Vol.108 (5), p.058301-058301, Article 058301</ispartof><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c438t-485466b7f94157f54d81304f62a7b86c9f8729efd7947cd5534d46fccdcc78453</citedby><cites>FETCH-LOGICAL-c438t-485466b7f94157f54d81304f62a7b86c9f8729efd7947cd5534d46fccdcc78453</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>230,314,780,784,885,27924,27925</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/22400967$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink><backlink>$$Uhttps://www.osti.gov/biblio/1099077$$D View this record in Osti.gov$$Hfree_for_read</backlink></links><search><creatorcontrib>Rupp, Matthias</creatorcontrib><creatorcontrib>Tkatchenko, Alexandre</creatorcontrib><creatorcontrib>Müller, Klaus-Robert</creatorcontrib><creatorcontrib>von Lilienfeld, O Anatole</creatorcontrib><title>Fast and accurate modeling of molecular atomization energies with machine learning</title><title>Physical review letters</title><addtitle>Phys Rev Lett</addtitle><description>We introduce a machine learning model to predict atomization energies of a diverse set of organic molecules, based on nuclear charges and atomic positions only. The problem of solving the molecular Schrödinger equation is mapped onto a nonlinear statistical regression problem of reduced complexity. Regression models are trained on and compared to atomization energies computed with hybrid density-functional theory. Cross validation over more than seven thousand organic molecules yields a mean absolute error of ∼10  kcal/mol. Applicability is demonstrated for the prediction of molecular atomization potential energy curves.</description><issn>0031-9007</issn><issn>1079-7114</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2012</creationdate><recordtype>article</recordtype><recordid>eNpNkF1rFTEQhoNY7Gn1L5TgjVd7nOxm83EpxVbhgKXo9ZIzO-mJ7CY1yVrqr-_KqcWrGYbnfQcexi4EbIWA7uPN4bHc0u8d1boVYLbQmw7EK7YRoG2jhZCv2QagE40F0KfsrJSfACBaZd6w07aVAFbpDbu9cqVyF0fuEJfsKvE5jTSFeMeTX_eJcJlc5q6mOfxxNaTIKVK-C1T4Q6gHPjs8hEh8IpfjmnvLTrybCr17nufsx9Xn75dfmt2366-Xn3YNys7URppeKrXX3krRa9_L0YgOpFet03uj0HqjW0t-1FZqHPu-k6NUHnFE1Eb23Tl7f-xNpYahYKiEB0wxEtZBgLWg9Qp9OEL3Of1aqNRhDgVpmlyktJTBtloZY1u5kupIYk6lZPLDfQ6zy49r1_DX-fCf8_VmhqPzNXjx_GLZzzS-xP5J7p4AgiKAMw</recordid><startdate>20120131</startdate><enddate>20120131</enddate><creator>Rupp, Matthias</creator><creator>Tkatchenko, Alexandre</creator><creator>Müller, Klaus-Robert</creator><creator>von Lilienfeld, O Anatole</creator><general>American Physical Society</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><scope>OTOTI</scope></search><sort><creationdate>20120131</creationdate><title>Fast and accurate modeling of molecular atomization energies with machine learning</title><author>Rupp, Matthias ; Tkatchenko, Alexandre ; Müller, Klaus-Robert ; von Lilienfeld, O Anatole</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c438t-485466b7f94157f54d81304f62a7b86c9f8729efd7947cd5534d46fccdcc78453</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2012</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Rupp, Matthias</creatorcontrib><creatorcontrib>Tkatchenko, Alexandre</creatorcontrib><creatorcontrib>Müller, Klaus-Robert</creatorcontrib><creatorcontrib>von Lilienfeld, O Anatole</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><collection>OSTI.GOV</collection><jtitle>Physical review letters</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Rupp, Matthias</au><au>Tkatchenko, Alexandre</au><au>Müller, Klaus-Robert</au><au>von Lilienfeld, O Anatole</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Fast and accurate modeling of molecular atomization energies with machine learning</atitle><jtitle>Physical review letters</jtitle><addtitle>Phys Rev Lett</addtitle><date>2012-01-31</date><risdate>2012</risdate><volume>108</volume><issue>5</issue><spage>058301</spage><epage>058301</epage><pages>058301-058301</pages><artnum>058301</artnum><issn>0031-9007</issn><eissn>1079-7114</eissn><abstract>We introduce a machine learning model to predict atomization energies of a diverse set of organic molecules, based on nuclear charges and atomic positions only. The problem of solving the molecular Schrödinger equation is mapped onto a nonlinear statistical regression problem of reduced complexity. Regression models are trained on and compared to atomization energies computed with hybrid density-functional theory. Cross validation over more than seven thousand organic molecules yields a mean absolute error of ∼10  kcal/mol. Applicability is demonstrated for the prediction of molecular atomization potential energy curves.</abstract><cop>United States</cop><pub>American Physical Society</pub><pmid>22400967</pmid><doi>10.1103/PhysRevLett.108.058301</doi><tpages>1</tpages><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 0031-9007
ispartof Physical review letters, 2012-01, Vol.108 (5), p.058301-058301, Article 058301
issn 0031-9007
1079-7114
language eng
recordid cdi_osti_scitechconnect_1099077
source American Physical Society:Jisc Collections:APS Read and Publish 2023-2025 (reading list)
title Fast and accurate modeling of molecular atomization energies with machine learning
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-02T19%3A35%3A31IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_osti_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Fast%20and%20accurate%20modeling%20of%20molecular%20atomization%20energies%20with%20machine%20learning&rft.jtitle=Physical%20review%20letters&rft.au=Rupp,%20Matthias&rft.date=2012-01-31&rft.volume=108&rft.issue=5&rft.spage=058301&rft.epage=058301&rft.pages=058301-058301&rft.artnum=058301&rft.issn=0031-9007&rft.eissn=1079-7114&rft_id=info:doi/10.1103/PhysRevLett.108.058301&rft_dat=%3Cproquest_osti_%3E927688924%3C/proquest_osti_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c438t-485466b7f94157f54d81304f62a7b86c9f8729efd7947cd5534d46fccdcc78453%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=927688924&rft_id=info:pmid/22400967&rfr_iscdi=true