Bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene
Abstract We present a way to dramatically accelerate Gaussian process models for interatomic force fields based on many-body kernels by mapping both forces and uncertainties onto functions of low-dimensional features. This allows for automated active learning of models combining near-quantum accurac...
Guardado en:
Autores principales: | , , , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
Nature Portfolio
2021
|
Materias: | |
Acceso en línea: | https://doaj.org/article/f04a0cef3d874031836d2287821283f8 |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:f04a0cef3d874031836d2287821283f8 |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:f04a0cef3d874031836d2287821283f82021-12-02T13:18:00ZBayesian force fields from active learning for simulation of inter-dimensional transformation of stanene10.1038/s41524-021-00510-y2057-3960https://doaj.org/article/f04a0cef3d874031836d2287821283f82021-03-01T00:00:00Zhttps://doi.org/10.1038/s41524-021-00510-yhttps://doaj.org/toc/2057-3960Abstract We present a way to dramatically accelerate Gaussian process models for interatomic force fields based on many-body kernels by mapping both forces and uncertainties onto functions of low-dimensional features. This allows for automated active learning of models combining near-quantum accuracy, built-in uncertainty, and constant cost of evaluation that is comparable to classical analytical models, capable of simulating millions of atoms. Using this approach, we perform large-scale molecular dynamics simulations of the stability of the stanene monolayer. We discover an unusual phase transformation mechanism of 2D stanene, where ripples lead to nucleation of bilayer defects, densification into a disordered multilayer structure, followed by formation of bulk liquid at high temperature or nucleation and growth of the 3D bcc crystal at low temperature. The presented method opens possibilities for rapid development of fast accurate uncertainty-aware models for simulating long-time large-scale dynamics of complex materials.Yu XieJonathan VandermauseLixin SunAndrea CepellottiBoris KozinskyNature PortfolioarticleMaterials of engineering and construction. Mechanics of materialsTA401-492Computer softwareQA76.75-76.765ENnpj Computational Materials, Vol 7, Iss 1, Pp 1-10 (2021) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
Materials of engineering and construction. Mechanics of materials TA401-492 Computer software QA76.75-76.765 |
spellingShingle |
Materials of engineering and construction. Mechanics of materials TA401-492 Computer software QA76.75-76.765 Yu Xie Jonathan Vandermause Lixin Sun Andrea Cepellotti Boris Kozinsky Bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene |
description |
Abstract We present a way to dramatically accelerate Gaussian process models for interatomic force fields based on many-body kernels by mapping both forces and uncertainties onto functions of low-dimensional features. This allows for automated active learning of models combining near-quantum accuracy, built-in uncertainty, and constant cost of evaluation that is comparable to classical analytical models, capable of simulating millions of atoms. Using this approach, we perform large-scale molecular dynamics simulations of the stability of the stanene monolayer. We discover an unusual phase transformation mechanism of 2D stanene, where ripples lead to nucleation of bilayer defects, densification into a disordered multilayer structure, followed by formation of bulk liquid at high temperature or nucleation and growth of the 3D bcc crystal at low temperature. The presented method opens possibilities for rapid development of fast accurate uncertainty-aware models for simulating long-time large-scale dynamics of complex materials. |
format |
article |
author |
Yu Xie Jonathan Vandermause Lixin Sun Andrea Cepellotti Boris Kozinsky |
author_facet |
Yu Xie Jonathan Vandermause Lixin Sun Andrea Cepellotti Boris Kozinsky |
author_sort |
Yu Xie |
title |
Bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene |
title_short |
Bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene |
title_full |
Bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene |
title_fullStr |
Bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene |
title_full_unstemmed |
Bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene |
title_sort |
bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene |
publisher |
Nature Portfolio |
publishDate |
2021 |
url |
https://doaj.org/article/f04a0cef3d874031836d2287821283f8 |
work_keys_str_mv |
AT yuxie bayesianforcefieldsfromactivelearningforsimulationofinterdimensionaltransformationofstanene AT jonathanvandermause bayesianforcefieldsfromactivelearningforsimulationofinterdimensionaltransformationofstanene AT lixinsun bayesianforcefieldsfromactivelearningforsimulationofinterdimensionaltransformationofstanene AT andreacepellotti bayesianforcefieldsfromactivelearningforsimulationofinterdimensionaltransformationofstanene AT boriskozinsky bayesianforcefieldsfromactivelearningforsimulationofinterdimensionaltransformationofstanene |
_version_ |
1718393364754202624 |