High-Dimensional Neural Network Potentials
in Chemistry, Physics and Materials Science
Theoretische Chemie, Institut für Physikalische Chemie,
Georg-August-Universität Göttingen, Germany
The reliability of the results obtained in computer simulations in chemistry, physics and materials science depends on the quality of the underlying potential-energy surface (PES). While the most accurate approach is to use electronic structure calculations like density-functional theory on-the-fly, the resulting ab initio molecular dynamics simulations are restricted to small systems and short simulation times. Consequently, a lot of effort has been invested for several decades in constructing more efficient atomistic potentials of varying form and complexity, which provide a direct functional relation between the atomic positions and the potential energy. Often these potentials are based on physical approximations, which necessarily reduce the accuracy of the PES.
In recent years a paradigm change has taken place by the introduction of machine learning (ML) potentials , which employ very flexible mathematical functions to represent a reference set of electronic structure data as accurately as possible. While the first ML potentials based on artificial neural networks have been proposed already in 1995 , early neural network potentials (NNPs) were only applicable to small systems containing a few degrees of freedom. Nowadays, ML potentials have become a practical tool for large-scale simulations based on three central concepts: the introduction of environment-dependent atomic energy contributions , the development of rotationally, translationally and permutation invariant descriptors , and a systematic way to build reference data sets for training NNPs . In this talk the current status of the method will be discussed, and some recent applications covering interfaces and bulk materials will be presented.
 J. Behler, J. Chem. Phys. 145 (2016) 170901.
 T. B. Blank, S. D. Brown, A. W. Calhoun, D. J. Doren, J. Chem. Phys. 103 (1995) 4129.
 J. Behler, M. Parrinello, Phys. Rev. Lett. 98 (2007) 146401.
 N. Artrith, J. Behler, Phys. Rev. B 85 (2012) 04543.