Universal Machine Learning Interatomic Potentials are Ready for Phonons
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912365499383808 |
|---|---|
| author | Loew, Antoine Sun, Dewen Wang, Hai-Chen Botti, Silvana Marques, Miguel A. L. |
| author_facet | Loew, Antoine Sun, Dewen Wang, Hai-Chen Botti, Silvana Marques, Miguel A. L. |
| contents | There has been an ongoing race for the past several years to develop the best universal machinelearning interatomic potential. This progress has led to increasingly accurate models for predictingenergy, forces, and stresses, combining innovative architectures with big data. Here, we benchmarkthese models on their ability to predict harmonic phonon properties, which are critical for under-standing the vibrational and thermal behavior of materials. Using around 10 000 ab initio phononcalculations, we evaluate model performance across various phonon-related parameters to test theuniversal applicability of these models. The results reveal that some models achieve high accuracyin predicting harmonic phonon properties. However, others still exhibit substantial inaccuracies,even if they excel in the prediction of the energy and the forces for materials close to dynamicalequilibrium. These findings highlight the importance of considering phonon-related properties inthe development of universal machine learning interatomic potentials. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_16551 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Universal Machine Learning Interatomic Potentials are Ready for Phonons Loew, Antoine Sun, Dewen Wang, Hai-Chen Botti, Silvana Marques, Miguel A. L. Materials Science Computational Physics There has been an ongoing race for the past several years to develop the best universal machinelearning interatomic potential. This progress has led to increasingly accurate models for predictingenergy, forces, and stresses, combining innovative architectures with big data. Here, we benchmarkthese models on their ability to predict harmonic phonon properties, which are critical for under-standing the vibrational and thermal behavior of materials. Using around 10 000 ab initio phononcalculations, we evaluate model performance across various phonon-related parameters to test theuniversal applicability of these models. The results reveal that some models achieve high accuracyin predicting harmonic phonon properties. However, others still exhibit substantial inaccuracies,even if they excel in the prediction of the energy and the forces for materials close to dynamicalequilibrium. These findings highlight the importance of considering phonon-related properties inthe development of universal machine learning interatomic potentials. |
| title | Universal Machine Learning Interatomic Potentials are Ready for Phonons |
| topic | Materials Science Computational Physics |
| url | https://arxiv.org/abs/2412.16551 |