Modeling of molecular atomization energies using machine learning
- Technical Univ. of Berlin (Germany). Machine Learning Group
- Max-Planck Society, Berlin (Germany). Fritz-Haber-Inst.
- Argonne National Lab. (ANL), Argonne, IL (United States)
Atomization energies are an important measure of chemical stability. Machine learning is used to model atomization energies of a diverse set of organic molecules, based on nuclear charges and atomic positions only. Our scheme maps the problem of solving the molecular time-independent Schrödinger equation onto a non-linear statistical regression problem. Kernel ridge regression models are trained on and compared to reference atomization energies computed using density functional theory (PBE0 approximation to Kohn-Sham level of theory). We use a diagonalized matrix representation of molecules based on the inter-nuclear Coulomb repulsion operator in conjunction with a Gaussian kernel. Validation on a set of over 7000 small organic molecules from the GDB database yields mean absolute error of ~10 kcal/mol, while reducing computational effort by several orders of magnitude. Applicability is demonstrated for prediction of binding energy curves using augmentation samples based on physical limits.
- Research Organization:
- Argonne National Laboratory (ANL), Argonne, IL (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC)
- OSTI ID:
- 1629374
- Journal Information:
- Journal of Cheminformatics, Vol. 4, Issue S1; Conference: 7.German Conference on Chemoinformatics: 25 CIC-Workshop, Goslar (Germany), 6-8 Nov 2011; ISSN 1758-2946
- Publisher:
- Chemistry Central Ltd.Copyright Statement
- Country of Publication:
- United States
- Language:
- English
Similar Records
Molecular-orbital-based machine learning for open-shell and multi-reference systems with kernel addition Gaussian process regression
Density functional theory based generalized effective fragment potential method