A systematic approach to generating accurate neural network potentials: the case of carbon
Abstract Availability of affordable and widely applicable interatomic potentials is the key needed to unlock the riches of modern materials modeling. Artificial neural network-based approaches for generating potentials are promising; however, neural network training requires large amounts of data, s...
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/3615bdf517374055a6e295aa00227d24 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:3615bdf517374055a6e295aa00227d24 |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:3615bdf517374055a6e295aa00227d242021-12-02T18:03:26ZA systematic approach to generating accurate neural network potentials: the case of carbon10.1038/s41524-021-00508-62057-3960https://doaj.org/article/3615bdf517374055a6e295aa00227d242021-04-01T00:00:00Zhttps://doi.org/10.1038/s41524-021-00508-6https://doaj.org/toc/2057-3960Abstract Availability of affordable and widely applicable interatomic potentials is the key needed to unlock the riches of modern materials modeling. Artificial neural network-based approaches for generating potentials are promising; however, neural network training requires large amounts of data, sampled adequately from an often unknown potential energy surface. Here we propose a self-consistent approach that is based on crystal structure prediction formalism and is guided by unsupervised data analysis, to construct an accurate, inexpensive, and transferable artificial neural network potential. Using this approach, we construct an interatomic potential for carbon and demonstrate its ability to reproduce first principles results on elastic and vibrational properties for diamond, graphite, and graphene, as well as energy ordering and structural properties of a wide range of crystalline and amorphous phases.Yusuf ShaiduEmine KüçükbenliRuggero LotFranco PellegriniEfthimios KaxirasStefano de GironcoliNature PortfolioarticleMaterials of engineering and construction. Mechanics of materialsTA401-492Computer softwareQA76.75-76.765ENnpj Computational Materials, Vol 7, Iss 1, Pp 1-13 (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 Yusuf Shaidu Emine Küçükbenli Ruggero Lot Franco Pellegrini Efthimios Kaxiras Stefano de Gironcoli A systematic approach to generating accurate neural network potentials: the case of carbon |
| description |
Abstract Availability of affordable and widely applicable interatomic potentials is the key needed to unlock the riches of modern materials modeling. Artificial neural network-based approaches for generating potentials are promising; however, neural network training requires large amounts of data, sampled adequately from an often unknown potential energy surface. Here we propose a self-consistent approach that is based on crystal structure prediction formalism and is guided by unsupervised data analysis, to construct an accurate, inexpensive, and transferable artificial neural network potential. Using this approach, we construct an interatomic potential for carbon and demonstrate its ability to reproduce first principles results on elastic and vibrational properties for diamond, graphite, and graphene, as well as energy ordering and structural properties of a wide range of crystalline and amorphous phases. |
| format |
article |
| author |
Yusuf Shaidu Emine Küçükbenli Ruggero Lot Franco Pellegrini Efthimios Kaxiras Stefano de Gironcoli |
| author_facet |
Yusuf Shaidu Emine Küçükbenli Ruggero Lot Franco Pellegrini Efthimios Kaxiras Stefano de Gironcoli |
| author_sort |
Yusuf Shaidu |
| title |
A systematic approach to generating accurate neural network potentials: the case of carbon |
| title_short |
A systematic approach to generating accurate neural network potentials: the case of carbon |
| title_full |
A systematic approach to generating accurate neural network potentials: the case of carbon |
| title_fullStr |
A systematic approach to generating accurate neural network potentials: the case of carbon |
| title_full_unstemmed |
A systematic approach to generating accurate neural network potentials: the case of carbon |
| title_sort |
systematic approach to generating accurate neural network potentials: the case of carbon |
| publisher |
Nature Portfolio |
| publishDate |
2021 |
| url |
https://doaj.org/article/3615bdf517374055a6e295aa00227d24 |
| work_keys_str_mv |
AT yusufshaidu asystematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT eminekucukbenli asystematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT ruggerolot asystematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT francopellegrini asystematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT efthimioskaxiras asystematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT stefanodegironcoli asystematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT yusufshaidu systematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT eminekucukbenli systematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT ruggerolot systematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT francopellegrini systematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT efthimioskaxiras systematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon AT stefanodegironcoli systematicapproachtogeneratingaccurateneuralnetworkpotentialsthecaseofcarbon |
| _version_ |
1718378744313282560 |