Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials

Abstract The universal mathematical form of machine-learning potentials (MLPs) shifts the core of development of interatomic potentials to collecting proper training data. Ideally, the training set should encompass diverse local atomic environments but conventional approaches are prone to sampling s...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Dongsun Yoo, Jisu Jung, Wonseok Jeong, Seungwu Han
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Materials of engineering and construction. Mechanics of materials TA401-492 Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/4d6bf771dd664f17a0d20898685ce07d
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4d6bf771dd664f17a0d20898685ce07d
record_format	dspace
spelling	oai:doaj.org-article:4d6bf771dd664f17a0d20898685ce07d2021-12-02T15:10:54ZMetadynamics sampling in atomic environment space for collecting training data for machine learning potentials10.1038/s41524-021-00595-52057-3960https://doaj.org/article/4d6bf771dd664f17a0d20898685ce07d2021-08-01T00:00:00Zhttps://doi.org/10.1038/s41524-021-00595-5https://doaj.org/toc/2057-3960Abstract The universal mathematical form of machine-learning potentials (MLPs) shifts the core of development of interatomic potentials to collecting proper training data. Ideally, the training set should encompass diverse local atomic environments but conventional approaches are prone to sampling similar configurations repeatedly, mainly due to the Boltzmann statistics. As such, practitioners handpick a large pool of distinct configurations manually, stretching the development period significantly. To overcome this hurdle, methods are being proposed that automatically generate training data. Herein, we suggest a sampling method optimized for gathering diverse yet relevant configurations semi-automatically. This is achieved by applying the metadynamics with the descriptor for the local atomic environment as a collective variable. As a result, the simulation is automatically steered toward unvisited local environment space such that each atom experiences diverse chemical environments without redundancy. We apply the proposed metadynamics sampling to H:Pt(111), GeTe, and Si systems. Throughout these examples, a small number of metadynamics trajectories can provide reference structures necessary for training high-fidelity MLPs. By proposing a semi-automatic sampling method tuned for MLPs, the present work paves the way to wider applications of MLPs to many challenging applications.Dongsun YooJisu JungWonseok JeongSeungwu HanNature PortfolioarticleMaterials of engineering and construction. Mechanics of materialsTA401-492Computer softwareQA76.75-76.765ENnpj Computational Materials, Vol 7, Iss 1, Pp 1-9 (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 Dongsun Yoo Jisu Jung Wonseok Jeong Seungwu Han Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials
description	Abstract The universal mathematical form of machine-learning potentials (MLPs) shifts the core of development of interatomic potentials to collecting proper training data. Ideally, the training set should encompass diverse local atomic environments but conventional approaches are prone to sampling similar configurations repeatedly, mainly due to the Boltzmann statistics. As such, practitioners handpick a large pool of distinct configurations manually, stretching the development period significantly. To overcome this hurdle, methods are being proposed that automatically generate training data. Herein, we suggest a sampling method optimized for gathering diverse yet relevant configurations semi-automatically. This is achieved by applying the metadynamics with the descriptor for the local atomic environment as a collective variable. As a result, the simulation is automatically steered toward unvisited local environment space such that each atom experiences diverse chemical environments without redundancy. We apply the proposed metadynamics sampling to H:Pt(111), GeTe, and Si systems. Throughout these examples, a small number of metadynamics trajectories can provide reference structures necessary for training high-fidelity MLPs. By proposing a semi-automatic sampling method tuned for MLPs, the present work paves the way to wider applications of MLPs to many challenging applications.
format	article
author	Dongsun Yoo Jisu Jung Wonseok Jeong Seungwu Han
author_facet	Dongsun Yoo Jisu Jung Wonseok Jeong Seungwu Han
author_sort	Dongsun Yoo
title	Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials
title_short	Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials
title_full	Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials
title_fullStr	Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials
title_full_unstemmed	Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials
title_sort	metadynamics sampling in atomic environment space for collecting training data for machine learning potentials
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/4d6bf771dd664f17a0d20898685ce07d
work_keys_str_mv	AT dongsunyoo metadynamicssamplinginatomicenvironmentspaceforcollectingtrainingdataformachinelearningpotentials AT jisujung metadynamicssamplinginatomicenvironmentspaceforcollectingtrainingdataformachinelearningpotentials AT wonseokjeong metadynamicssamplinginatomicenvironmentspaceforcollectingtrainingdataformachinelearningpotentials AT seungwuhan metadynamicssamplinginatomicenvironmentspaceforcollectingtrainingdataformachinelearningpotentials
_version_	1718387626583523328

Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials

Ejemplares similares