[en] 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 Schrodinger 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 similar to 10 kcal/mol. Applicability is demonstrated for the prediction of molecular atomization potential energy curves.
Disciplines :
Physics
Author, co-author :
Rupp, Matthias; Machine Learning Group, Technical University of Berlin, Franklinstr 28/29, 10587 Berlin, Germany ; Institute of Pure and Applied Mathematics, University of California Los Angeles, Los Angeles, California 90095, USA
Tkatchenko, Alexandre ; Institute of Pure and Applied Mathematics, University of California Los Angeles, Los Angeles, California 90095, USA ; Fritz-Haber-Institut der Max-Planck-Gesellschaft, 14195 Berlin, Germany
Mueller, Klaus-Robert; Machine Learning Group, Technical University of Berlin, Franklinstr 28/29, 10587 Berlin, Germany ; Institute of Pure and Applied Mathematics, University of California Los Angeles, Los Angeles, California 90095, USA
von Lilienfeld, O. Anatole; Institute of Pure and Applied Mathematics, University of California Los Angeles, Los Angeles, California 90095, USA ; Argonne Leadership Computing Facility, Argonne National Laboratory, Argonne, Illinois 60439, USA
External co-authors :
yes
Title :
Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
Publication date :
2012
Journal title :
PHYSICAL REVIEW LETTERS
ISSN :
0031-9007
Publisher :
AMER PHYSICAL SOC, ONE PHYSICS ELLIPSE, COLLEGE PK, MD 20740-3844 USA, Unknown/unspecified
