Back to Search
Start Over
Molecular force fields with gradient-domain machine learning (GDML): Comparison and synergies with classical force fields
- Source :
- The Journal of chemical physics. 153(12)
- Publication Year :
- 2020
-
Abstract
- This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This article appeared in J. Chem. Phys. 153, 124109 (2020) and may be found at https://doi.org/10.1063/5.0023005.<br />Modern machine learning force fields (ML-FF) are able to yield energy and force predictions at the accuracy of high-level ab initio methods, but at a much lower computational cost. On the other hand, classical molecular mechanics force fields (MM-FF) employ fixed functional forms and tend to be less accurate, but considerably faster and transferable between molecules of the same class. In this work, we investigate how both approaches can complement each other. We contrast the ability of ML-FF for reconstructing dynamic and thermodynamic observables to MM-FFs in order to gain a qualitative understanding of the differences between the two approaches. This analysis enables us to modify the generalized AMBER force field by reparametrizing short-range and bonded interactions with more expressive terms to make them more accurate, without sacrificing the key properties that make MM-FFs so successful.
- Subjects :
- Computer science
force fields
Physics [G04] [Physical, chemical, mathematical & earth Sciences]
Ab initio
FOS: Physical sciences
General Physics and Astronomy
ML-FF
Machine learning
computer.software_genre
01 natural sciences
Molecular mechanics
Force field (chemistry)
Physics - Chemical Physics
0103 physical sciences
541 Physikalische Chemie
Physics - Atomic and Molecular Clusters
Physics - Biological Physics
Physical and Theoretical Chemistry
010306 general physics
004 Datenverarbeitung
Informatik
Chemical Physics (physics.chem-ph)
010304 chemical physics
business.industry
Observable
MM-FF
Computational Physics (physics.comp-ph)
molecular mechanics
machine learning
Physique [G04] [Physique, chimie, mathématiques & sciences de la terre]
Biological Physics (physics.bio-ph)
Physics - Data Analysis, Statistics and Probability
ddc:541
Artificial intelligence
ddc:004
Atomic and Molecular Clusters (physics.atm-clus)
business
Physics - Computational Physics
computer
Data Analysis, Statistics and Probability (physics.data-an)
Subjects
Details
- ISSN :
- 10897690
- Volume :
- 153
- Issue :
- 12
- Database :
- OpenAIRE
- Journal :
- The Journal of chemical physics
- Accession number :
- edsair.doi.dedup.....e76cfb14c1ea7eb27d98821242052770