Convolutional neural networks for atomistic systems
| Autor: | Isaac Tamblyn, Kyle Mills, Iryna Luchak, Kevin Ryczko, Christa M. Homenick |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
General Computer Science
Extrapolation FOS: Physical sciences General Physics and Astronomy 02 engineering and technology 01 natural sciences Convolutional neural network Ab initio quantum chemistry methods Approximation error convolutional neural networks 0103 physical sciences Physics::Atomic and Molecular Clusters General Materials Science Statistical physics 010306 general physics density functional theory Physics Condensed Matter - Materials Science Artificial neural network business.industry Deep learning deep learning Materials Science (cond-mat.mtrl-sci) General Chemistry 2D materials 021001 nanoscience & nanotechnology Computational Mathematics dimer molecules Mechanics of Materials Density functional theory Artificial intelligence 0210 nano-technology business Interpolation |
| Zdroj: | Computational Materials Science. 149:134-142 |
| ISSN: | 0927-0256 |
| Popis: | We introduce a new method, called CNNAS (convolutional neural networks for atomistic systems), for calculating the total energy of atomic systems which rivals the computational cost of empirical potentials while maintaining the accuracy of \emph{ab initio} calculations. This method uses deep convolutional neural networks (CNNs), where the input to these networks are simple representations of the atomic structure. We use this approach to predict energies obtained using density functional theory (DFT) for 2D hexagonal lattices of various types. Using a dataset consisting of graphene, hexagonal boron nitride (hBN), and graphene-hBN heterostructures, with and without defects, we trained a deep CNN that is capable of predicting DFT energies to an extremely high accuracy, with a mean absolute error (MAE) of 0.198 meV / atom (maximum absolute error of 16.1 meV / atom). To explore our new methodology, we investigate the ability of a deep neural network (DNN) in predicting a Lennard-Jones energy and separation distance for a dataset of dimer molecules in both two and three dimensions. In addition, we systematically investigate the flexibility of the deep learning models by performing interpolation and extrapolation tests. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect