Generalized Density-Functional Tight-Binding Repulsive Potentials from Unsupervised Machine Learning
Authors:
- Julian J. KranzJKJulian J. Kranz
- Maximilian KubillusMKMaximilian Kubillus
- O. Anatole von LilienfeldOLO. Anatole von Lilienfeld
- Marcus ElstnerMEMarcus Elstner
- 26 March 2018
- journal article
- research article
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 14 (5) , 2341-2352
Abstract
We combine the approximate Density-Functional Tight-Binding (DFTB) method with unsupervised machine learning . This allows to improve transferability and accuracy, make use of large quantum chemical data sets for the parametrization, and to efficiently automatize the parametrization process of DFTB. For this purpose, generalized pair-potentials are introduced, where the chemical environmental is included during the learning process leading to more specific effective two-body potentials. We train on energies and forces of equilibrium and non-equilibrium structures of 2100 molecules, and test on ∼ 130.000 organic molecules containing O, N, C, H and F atoms. Atomization energies of the reference method can be reproduced within an error of ∼ 2.6 kcal/mol, indicating drastic improvement over standard DFTB.Keywords
Funding Information
- Schweizerischer Nationalfonds zur F?rderung der Wissenschaftlichen Forschung (PP00P2_138932)
- Deutsche Forschungsgemeinschaft (SFB 1249)
This publication has 42 references indexed in Scilit:
- Enumeration of 166 Billion Organic Small Molecules in the Chemical Universe Database GDB-17Journal of Chemical Information and Modeling, 2012
- DFTB3: Extension of the Self-Consistent-Charge Density-Functional Tight-Binding Method (SCC-DFTB)Journal of Chemical Theory and Computation, 2011
- Comparison of SCC-DFTB and NDDO-Based Semiempirical Molecular Orbital Methods for Organic MoleculesThe Journal of Physical Chemistry A, 2006
- Construction of tight-binding-like potentials on the basis of density-functional theory: Application to carbonPhysical Review B, 1995
- Mean shift, mode seeking, and clusteringIEEE Transactions on Pattern Analysis and Machine Intelligence, 1995
- Development of the Colle-Salvetti correlation-energy formula into a functional of the electron densityPhysical Review B, 1988
- The estimation of the gradient of a density function, with applications in pattern recognitionIEEE Transactions on Information Theory, 1975
- Self-Consistent Equations Including Exchange and Correlation EffectsPhysical Review B, 1965
- Inhomogeneous Electron GasPhysical Review B, 1964
- Analysis of a complex of statistical variables into principal components.Journal of Educational Psychology, 1933