Managing uncertainty in data-derived densities to accelerate density functional theory
| Autor: | Fowler, AT, Pickard, CJ, Elliott, JA |
|---|---|
| Přispěvatelé: | Fowler, AT [0000-0002-7360-3078], Pickard, CJ [0000-0002-9684-5432], Elliott, JA [0000-0002-4887-6250], Apollo - University of Cambridge Repository |
| Rok vydání: | 2019 |
| Předmět: |
Electron density
non-Bayesian method FOS: Physical sciences 02 engineering and technology 01 natural sciences Acceleration Quality (physics) 0103 physical sciences General Materials Science Limit (mathematics) Statistical physics electron density 010306 general physics density functional theory Mathematics Condensed Matter - Materials Science Materials Science (cond-mat.mtrl-sci) Sampling (statistics) Computational Physics (physics.comp-ph) 021001 nanoscience & nanotechnology Condensed Matter Physics Atomic and Molecular Physics and Optics Regression parametric regression machine learning Density functional theory 0210 nano-technology Ground state Physics - Computational Physics |
| Zdroj: | Journal of Physics: Materials. 2:034001 |
| ISSN: | 2515-7639 |
| DOI: | 10.1088/2515-7639/ab0b4a |
| Popis: | Faithful representations of atomic environments and general models for regression can be harnessed to learn electron densities that are close to the ground state. One of the applications of data-derived electron densities is to orbital-free density functional theory. However, extrapolations of densities learned from a training set to dissimilar structures could result in inaccurate results, which would limit the applicability of the method. Here, we show that a non-Bayesian approach can produce estimates of uncertainty which can successfully distinguish accurate from inaccurate predictions of electron density. We apply our approach to density functional theory where we initialise calculations with data-derived densities only when we are confident about their quality. This results in a guaranteed acceleration to self-consistency for configurations that are similar to those seen during training and could be useful for sampling based methods, where previous ground state densities cannot be used to initialise subsequent calculations. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|