English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Machine learning of accurate energy-conserving molecular force fields

Chmiela, S., Tkatchenko, A., Sauceda, H. E., Poltavsky, I., Schütt, K. T., & Müller, K.-R. (2017). Machine learning of accurate energy-conserving molecular force fields. Science Advances, 3(5): e1603015. doi:10.1126/sciadv.1603015.

Item is

Files

show Files
hide Files
:
chmiela.pdf (Publisher version), 2MB
Name:
chmiela.pdf
Description:
-
OA-Status:
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
2017
Copyright Info:
© The Authors

Locators

show

Creators

show
hide
 Creators:
Chmiela, Stefan1, Author
Tkatchenko, Alexandre2, 3, Author           
Sauceda, Huziel E.2, Author           
Poltavsky, Igor3, Author
Schütt, Kristof T.1, Author
Müller, Klaus-Robert1, 4, Author
Affiliations:
1Machine Learning Group, Technische Universität Berlin, 10587 Berlin, Germany, ou_persistent22              
2Theory, Fritz Haber Institute, Max Planck Society, ou_634547              
3Physics and Materials Science Research Unit, University of Luxembourg, L-1511 Luxembourg, ou_persistent22              
4Department of Brain and Cognitive Engineering, Korea University, Anam-dong, Seongbuk-gu, Seoul 136-713, Korea, ou_persistent22              

Content

show
hide
Free keywords: -
 Abstract: Using 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 potentialenergy surfaces of intermediate-size 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.

Details

show
hide
Language(s): eng - English
 Dates: 2016-12-012017-03-072017-05-05
 Publication Status: Published online
 Pages: 7
 Publishing info: -
 Table of Contents: -
 Rev. Type: Peer
 Identifiers: DOI: 10.1126/sciadv.1603015
 Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show
hide
Title: Science Advances
  Other : Sci. Adv.
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: Washington : AAAS
Pages: 7 Volume / Issue: 3 (5) Sequence Number: e1603015 Start / End Page: - Identifier: Other: 2375-2548
CoNE: https://pure.mpg.de/cone/journals/resource/2375-2548