Skip to Main content Skip to Navigation
Journal articles

Measuring transferability issues in machine-learning force fields: The example of Gold-Iron interactions with linearized potentials

Abstract : Machine-learning force fields have been increasingly employed in order to extend the possibility of current first-principles calculations. However, the transferability of the obtained potential can not always be guaranteed in situations that are outside the original database. To study such limitation, we examined the very difficult case of the interactions in gold-iron nanoparticles. For the machinelearning potential, we employed a linearized formulation that is parameterized using a penalizing regression scheme which allows us to control the complexity of the obtained potential. We showed that while having a more complex potential allows for a better agreement with the training database, it can also lead to overfitting issues and a lower accuracy in untrained systems.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03052129
Contributor : Julien Lam <>
Submitted on : Thursday, December 10, 2020 - 2:54:13 PM
Last modification on : Tuesday, April 6, 2021 - 5:14:05 PM
Long-term archiving on: : Thursday, March 11, 2021 - 7:43:56 PM

File

1912.10761.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03052129, version 1

Citation

Magali Benoit, Jonathan Amodeo, Ségolène Combettes, Ibrahim Khaled, Aurélien Roux, et al.. Measuring transferability issues in machine-learning force fields: The example of Gold-Iron interactions with linearized potentials. Machine Learning: Science and Technology, 2020. ⟨hal-03052129⟩

Share

Metrics

Record views

117

Files downloads

45