EUVIMEDEUVIMED entdecken
Crossref

How to train a neural network potential

Alea Miako Tokita iD, Jörg Behler iD

The Journal of Chemical Physics · 28.09.2023 · 159 (12)

DOI: 10.1063/5.0160326 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of Chemical Physics, online erschienen am 27.09.2023. Verfasst von Alea Miako Tokita und 1 weiteren Autor:innen. Beteiligte Einrichtung: Lehrstuhl für Theoretische Chemie II, Ruhr-Universität Bochum, 44780 Bochum, Germany and Research Center Chemical Sciences and Sustainability, Resear…. Die Arbeit wurde bisher 84-mal zitiert (Crossref) und nennt 155 Literaturangaben. Gefördert von: Deutsche Forschungsgemeinschaft.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The introduction of modern Machine Learning Potentials (MLPs) has led to a paradigm change in the development of potential energy surfaces for atomistic simulations. By providing efficient access to energies and forces, they allow us to perform large-scale simulations of extended systems, which are not directly accessible by demanding first-principles methods. In these simulations, MLPs can reach the accuracy of electronic structure calculations, provided that they have been properly trained and validated using a suitable set of reference data. Due to their highly flexible functional form, the construction of MLPs has to be done with great care. In this Tutorial, we describe the necessary key steps for training reliable MLPs, from data generation via training to final validation. The procedure, which is illustrated for the example of a high-dimensional neural network potential, is general and applicable to many types of MLPs.

Autor:innen und Institutionen

  • Lehrstuhl für Theoretische Chemie II, Ruhr-Universität Bochum, 44780 Bochum, Germany and Research Center Chemical Sciences and Sustainability, Research Alliance Ruhr, 44780 Bochum, Germany.

ORCID-Profile: Alea Miako Tokita, Jörg Behler

Förderung

  • Deutsche Forschungsgemeinschaft

Zitieren

APA

Tokita, A. M., & Behler, J. (2023). How to train a neural network potential. The Journal of Chemical Physics, 159(12). https://doi.org/10.1063/5.0160326

Vancouver

Tokita AM, Behler J. How to train a neural network potential. The Journal of Chemical Physics. 2023;159(12). doi:10.1063/5.0160326

BibTeX
@article{Tokita2023,
  author = {Tokita, Alea Miako and Behler, Jörg},
  title = {How to train a neural network potential},
  journal = {The Journal of Chemical Physics},
  year = {2023},
  volume = {159},
  number = {12},
  doi = {10.1063/5.0160326}
}
Kostenlos starten →