Machine learning of accurate energy-conserving molecular force fields

dc.contributor.authorChmiela, Stefan
dc.contributor.authorTkatchenko, Alexandre
dc.contributor.authorSauceda, Huziel E.
dc.contributor.authorPoltavsky, Igor
dc.contributor.authorSchütt, Kristof T.
dc.contributor.authorMüller, Klaus-Robert
dc.date.accessioned2018-04-19T08:50:36Z
dc.date.available2018-04-19T08:50:36Z
dc.date.issued2017
dc.description.abstractUsing conservation of energy - a fundamental property of closed classical and quantum mechanical systems - we develop an efficient gradient-domain machine learning (GDML) approach to construct accurate molecular force fields using a restricted number of samples from ab initio molecular dynamics (AIMD) trajectories. The GDML implementation is able to reproduce global potential energy surfaces of intermediate-sized molecules with an accuracy of 0.3 kcal mol−1 for energies and 1 kcal mol−1 Å̊−1 for atomic forces using only 1000 conformational geometries for training. We demonstrate this accuracy for AIMD trajectories of molecules, including benzene, toluene, naphthalene, ethanol, uracil, and aspirin. The challenge of constructing conservative force fields is accomplished in our work by learning in a Hilbert space of vector-valued functions that obey the law of energy conservation. The GDML approach enables quantitative molecular dynamics simulations for molecules at a fraction of cost of explicit AIMD calculations, thereby allowing the construction of efficient force fields with the accuracy and transferability of high-level ab initio methods.en
dc.description.sponsorshipBMBF, 01IS14013A, Verbundprojekt: BBDC - Berliner Kompetenzzentrum für Big Dataen
dc.description.sponsorshipDFG, MU 987/20-1en
dc.identifier.eissn2375-2548
dc.identifier.issn2375-2548
dc.identifier.urihttps://depositonce.tu-berlin.de//handle/11303/7659
dc.identifier.urihttp://dx.doi.org/10.14279/depositonce-6849
dc.language.isoen
dc.rights.urihttps://creativecommons.org/licenses/by-nc/4.0/
dc.subject.ddc500 Naturwissenschaften und Mathematik
dc.subject.otherforce fielden
dc.subject.othermachine learningen
dc.subject.othergradient fielden
dc.subject.otherpotential-energy surfaceen
dc.subject.otherenergy conservationen
dc.subject.otherkernel regressionen
dc.subject.otherpath integralsen
dc.subject.othermolecular dynamicsen
dc.titleMachine learning of accurate energy-conserving molecular force fieldsen
dc.typeArticle
dc.type.versionpublishedVersionen
dcterms.bibliographicCitation.articlenumbere1603015
dcterms.bibliographicCitation.doi10.1126/sciadv.1603015
dcterms.bibliographicCitation.issue5
dcterms.bibliographicCitation.journaltitleScience Advancesen
dcterms.bibliographicCitation.originalpublishernameAmerican Association for the Advancement of Science (AAAS)
dcterms.bibliographicCitation.originalpublisherplaceWashington, DC [u.a.]
dcterms.bibliographicCitation.volume3
tub.accessrights.dnbfree
tub.affiliationFak. 4 Elektrotechnik und Informatik>Inst. Softwaretechnik und Theoretische Informatik>FG Maschinelles Lernende
tub.affiliation.facultyFak. 4 Elektrotechnik und Informatikde
tub.affiliation.groupFG Maschinelles Lernende
tub.affiliation.instituteInst. Softwaretechnik und Theoretische Informatikde
tub.publisher.universityorinstitutionTechnische Universität Berlinde
Files
Original bundle
Now showing 1 - 1 of 1
Loading…
Thumbnail Image
Name:
e1603015.full.pdf
Size:
1.15 MB
Format:
Adobe Portable Document Format
Description:
Collections