Open Access
Open access
volume 108 issue 5 publication number 058301

Fast and accurate modeling of molecular atomization energies with machine learning.

Publication typeJournal Article
Publication date2012-01-31
scimago Q1
wos Q1
SJR2.856
CiteScore15.6
Impact factor9.0
ISSN00319007, 10797114
General Physics and Astronomy
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.
Found 
Found 

Top-30

Journals

50
100
150
200
250
Journal of Chemical Physics
202 publications, 10.73%
Journal of Chemical Theory and Computation
127 publications, 6.74%
Journal of Chemical Information and Modeling
65 publications, 3.45%
Machine Learning: Science and Technology
62 publications, 3.29%
Journal of Physical Chemistry Letters
55 publications, 2.92%
Physical Review B
52 publications, 2.76%
Journal of Physical Chemistry A
50 publications, 2.66%
npj Computational Materials
48 publications, 2.55%
Physical Chemistry Chemical Physics
46 publications, 2.44%
Computational Materials Science
45 publications, 2.39%
Nature Communications
41 publications, 2.18%
Chemical Science
35 publications, 1.86%
Physical Review Materials
25 publications, 1.33%
Journal of Physical Chemistry C
23 publications, 1.22%
Journal of Computational Chemistry
23 publications, 1.22%
Scientific data
22 publications, 1.17%
Digital Discovery
21 publications, 1.12%
Scientific Reports
20 publications, 1.06%
International Journal of Quantum Chemistry
17 publications, 0.9%
Journal of Physical Chemistry B
16 publications, 0.85%
Physical Review Letters
16 publications, 0.85%
Chemical Reviews
16 publications, 0.85%
Lecture Notes in Physics
15 publications, 0.8%
Chemistry of Materials
12 publications, 0.64%
Computer Physics Communications
12 publications, 0.64%
ACS Omega
12 publications, 0.64%
Physical Review E
10 publications, 0.53%
Journal of Cheminformatics
10 publications, 0.53%
Wiley Interdisciplinary Reviews: Computational Molecular Science
10 publications, 0.53%
50
100
150
200
250

Publishers

50
100
150
200
250
300
350
400
450
American Chemical Society (ACS)
430 publications, 22.84%
Springer Nature
271 publications, 14.39%
AIP Publishing
232 publications, 12.32%
Elsevier
218 publications, 11.58%
Royal Society of Chemistry (RSC)
149 publications, 7.91%
Wiley
136 publications, 7.22%
American Physical Society (APS)
115 publications, 6.11%
IOP Publishing
108 publications, 5.74%
MDPI
37 publications, 1.96%
Institute of Electrical and Electronics Engineers (IEEE)
35 publications, 1.86%
Taylor & Francis
12 publications, 0.64%
Frontiers Media S.A.
11 publications, 0.58%
American Association for the Advancement of Science (AAAS)
10 publications, 0.53%
Cold Spring Harbor Laboratory
9 publications, 0.48%
Proceedings of the National Academy of Sciences (PNAS)
6 publications, 0.32%
Annual Reviews
6 publications, 0.32%
Oxford University Press
5 publications, 0.27%
Public Library of Science (PLoS)
5 publications, 0.27%
Association for Computing Machinery (ACM)
4 publications, 0.21%
Science in China Press
4 publications, 0.21%
Japan Society of Applied Physics
3 publications, 0.16%
American Astronomical Society
3 publications, 0.16%
Cambridge University Press
3 publications, 0.16%
Walter de Gruyter
3 publications, 0.16%
Bentham Science Publishers Ltd.
2 publications, 0.11%
Copernicus
2 publications, 0.11%
Society for Industrial and Applied Mathematics (SIAM)
2 publications, 0.11%
Japan Institute of Metals
2 publications, 0.11%
Physical Society of Japan
2 publications, 0.11%
50
100
150
200
250
300
350
400
450
  • We do not take into account publications without a DOI.
  • Statistics recalculated weekly.

Are you a researcher?

Create a profile to get free access to personal recommendations for colleagues and new articles.
Metrics
1.9k
Share
Cite this
GOST |
Cite this
GOST Copy
Rupp M. et al. Fast and accurate modeling of molecular atomization energies with machine learning. // Physical Review Letters. 2012. Vol. 108. No. 5. 058301
GOST all authors (up to 50) Copy
Rupp M., Tkatchenko A., MÜLLER K., Muller K., Von Lilienfeld O. A., von Lilienfeld O. A. Fast and accurate modeling of molecular atomization energies with machine learning. // Physical Review Letters. 2012. Vol. 108. No. 5. 058301
RIS |
Cite this
RIS Copy
TY - JOUR
DO - 10.1103/physrevlett.108.058301
UR - https://doi.org/10.1103/physrevlett.108.058301
TI - Fast and accurate modeling of molecular atomization energies with machine learning.
T2 - Physical Review Letters
AU - Rupp, Matthias
AU - Tkatchenko, Alexandre
AU - MÜLLER, KLAUS-ROBERT
AU - Muller, Klause
AU - Von Lilienfeld, O Anatole
AU - von Lilienfeld, O. Anatole
PY - 2012
DA - 2012/01/31
PB - American Physical Society (APS)
IS - 5
VL - 108
PMID - 22400967
SN - 0031-9007
SN - 1079-7114
ER -
BibTex
Cite this
BibTex (up to 50 authors) Copy
@article{2012_Rupp,
author = {Matthias Rupp and Alexandre Tkatchenko and KLAUS-ROBERT MÜLLER and Klause Muller and O Anatole Von Lilienfeld and O. Anatole von Lilienfeld},
title = {Fast and accurate modeling of molecular atomization energies with machine learning.},
journal = {Physical Review Letters},
year = {2012},
volume = {108},
publisher = {American Physical Society (APS)},
month = {jan},
url = {https://doi.org/10.1103/physrevlett.108.058301},
number = {5},
pages = {058301},
doi = {10.1103/physrevlett.108.058301}
}