Latent Ewald summation for machine learning of long-range interactions
Cheng B. 2025. Latent Ewald summation for machine learning of long-range interactions. npj Computational Materials. 11, 80.
Download
Journal Article
| Published
| English
Scopus indexed
Author
Corresponding author has ISTA affiliation
Department
Abstract
Machine learning interatomic potentials (MLIPs) often neglect long-range interactions, such as electrostatic and dispersion forces. In this work, we introduce a straightforward and efficient method to account for long-range interactions by learning a hidden variable from local atomic descriptors and applying an Ewald summation to this variable. We demonstrate that in systems including charged and polar molecular dimers, bulk water, and water-vapor interface, standard short-ranged MLIPs can lead to unphysical predictions even when employing message passing. The long-range models effectively eliminate these artifacts, with only about twice the computational cost of short-range MLIPs.
Publishing Year
Date Published
2025-03-26
Journal Title
npj Computational Materials
Publisher
Springer Nature
Acknowledgement
B. C. thanks David Limmer for providing the water slab dataset, and Carolin Faller for the NaCl dataset.
Volume
11
Article Number
80
eISSN
IST-REx-ID
Cite this
Cheng B. Latent Ewald summation for machine learning of long-range interactions. npj Computational Materials. 2025;11. doi:10.1038/s41524-025-01577-7
Cheng, B. (2025). Latent Ewald summation for machine learning of long-range interactions. Npj Computational Materials. Springer Nature. https://doi.org/10.1038/s41524-025-01577-7
Cheng, Bingqing. “Latent Ewald Summation for Machine Learning of Long-Range Interactions.” Npj Computational Materials. Springer Nature, 2025. https://doi.org/10.1038/s41524-025-01577-7.
B. Cheng, “Latent Ewald summation for machine learning of long-range interactions,” npj Computational Materials, vol. 11. Springer Nature, 2025.
Cheng B. 2025. Latent Ewald summation for machine learning of long-range interactions. npj Computational Materials. 11, 80.
Cheng, Bingqing. “Latent Ewald Summation for Machine Learning of Long-Range Interactions.” Npj Computational Materials, vol. 11, 80, Springer Nature, 2025, doi:10.1038/s41524-025-01577-7.
All files available under the following license(s):
Creative Commons Attribution 4.0 International Public License (CC-BY 4.0):
Main File(s)
File Name
2025_npjCompMaterials_Cheng.pdf
1.61 MB
Access Level

Date Uploaded
2025-04-08
MD5 Checksum
cc99b7407a12139d9b2d8457961935ae
Export
Marked PublicationsOpen Data ISTA Research Explorer
Sources
arXiv 2408.15165