Riemannian Denoising Model for Molecular Structure Optimization with Chemical Accuracy
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914261727444992 |
|---|---|
| author | Woo, Jeheon Kim, Seonghwan Kim, Jun Hyeong Kim, Woo Youn |
| author_facet | Woo, Jeheon Kim, Seonghwan Kim, Jun Hyeong Kim, Woo Youn |
| contents | We introduce a framework for molecular structure optimization using denoising model on a physics-informed Riemannian manifold (R-DM). Unlike conventional approaches operating in Euclidean space, our method leverages a Riemannian metric that better aligns with molecular energy change, enabling more robust modeling of potential energy surfaces. By incorporating internal coordinates reflective of energetic properties, R-DM achieves chemical accuracy with an energy error below 1 kcal/mol. Comparative evaluations on QM9, QM7-X, and GEOM datasets demonstrate improvements in both structural and energetic accuracy, surpassing conventional Euclidean-based denoising models. This approach highlights the potential of physics-informed coordinates for tackling complex molecular optimization problems, with implications for tasks in computational chemistry and materials science. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_19769 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Riemannian Denoising Model for Molecular Structure Optimization with Chemical Accuracy Woo, Jeheon Kim, Seonghwan Kim, Jun Hyeong Kim, Woo Youn Machine Learning Chemical Physics We introduce a framework for molecular structure optimization using denoising model on a physics-informed Riemannian manifold (R-DM). Unlike conventional approaches operating in Euclidean space, our method leverages a Riemannian metric that better aligns with molecular energy change, enabling more robust modeling of potential energy surfaces. By incorporating internal coordinates reflective of energetic properties, R-DM achieves chemical accuracy with an energy error below 1 kcal/mol. Comparative evaluations on QM9, QM7-X, and GEOM datasets demonstrate improvements in both structural and energetic accuracy, surpassing conventional Euclidean-based denoising models. This approach highlights the potential of physics-informed coordinates for tackling complex molecular optimization problems, with implications for tasks in computational chemistry and materials science. |
| title | Riemannian Denoising Model for Molecular Structure Optimization with Chemical Accuracy |
| topic | Machine Learning Chemical Physics |
| url | https://arxiv.org/abs/2411.19769 |