An automated approach for developing neural network interatomic potentials with FLAME

H. Mirhosseini, H. Tahmasbi, S.R. Kuchana, A. Ghasemi, T. Kühne, Comput. Mater. Sci. 197 (2021) 110567.

Download
No fulltext has been uploaded.
Journal Article | English
Author
Abstract
The performance of machine learning interatomic potentials relies on the quality of the training dataset. In this work, we present an approach for generating diverse and representative training data points which initiates with ab initio calculations for bulk structures. The data generation and potential construction further proceed side-by-side in a cyclic process of training the neural network and crystal structure prediction based on the developed interatomic potentials. All steps of the data generation and potential development are performed with minimal human intervention. We show the reliability of our approach by assessing the performance of neural network potentials developed for two inorganic systems.
Publishing Year
Journal Title
Comput. Mater. Sci.
Volume
197
Page
110567
ISSN
LibreCat-ID

Cite this

Mirhosseini H, Tahmasbi H, Kuchana SR, Ghasemi A, Kühne T. An automated approach for developing neural network interatomic potentials with FLAME. Comput Mater Sci. 2021;197:110567. doi:10.1016/j.commatsci.2021.110567
Mirhosseini, H., Tahmasbi, H., Kuchana, S. R., Ghasemi, A., & Kühne, T. (2021). An automated approach for developing neural network interatomic potentials with FLAME. Comput. Mater. Sci., 197, 110567. https://doi.org/10.1016/j.commatsci.2021.110567
@article{Mirhosseini_Tahmasbi_Kuchana_Ghasemi_Kühne_2021, title={An automated approach for developing neural network interatomic potentials with FLAME}, volume={197}, DOI={10.1016/j.commatsci.2021.110567}, journal={Comput. Mater. Sci.}, publisher={Elsevier}, author={Mirhosseini, Hossein and Tahmasbi, Hossein and Kuchana, Sai Ram and Ghasemi, Alireza and Kühne, Thomas}, year={2021}, pages={110567} }
Mirhosseini, Hossein, Hossein Tahmasbi, Sai Ram Kuchana, Alireza Ghasemi, and Thomas Kühne. “An Automated Approach for Developing Neural Network Interatomic Potentials with FLAME.” Comput. Mater. Sci. 197 (2021): 110567. https://doi.org/10.1016/j.commatsci.2021.110567.
H. Mirhosseini, H. Tahmasbi, S. R. Kuchana, A. Ghasemi, and T. Kühne, “An automated approach for developing neural network interatomic potentials with FLAME,” Comput. Mater. Sci., vol. 197, p. 110567, 2021, doi: 10.1016/j.commatsci.2021.110567.
Mirhosseini, Hossein, et al. “An Automated Approach for Developing Neural Network Interatomic Potentials with FLAME.” Comput. Mater. Sci., vol. 197, Elsevier, 2021, p. 110567, doi:10.1016/j.commatsci.2021.110567.

Export

Marked Publications

Open Data LibreCat

Search this title in

Google Scholar