Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2402.12335 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913962890625024 |
|---|---|
| author | Li, Chenghan Sharir, Or Yuan, Shunyue Chan, Garnet K. |
| author_facet | Li, Chenghan Sharir, Or Yuan, Shunyue Chan, Garnet K. |
| contents | Drawing inspiration from the domain of image super-resolution, we view the electron density as a 3D grayscale image and use a convolutional residual network to transform a crude and trivially generated guess of the molecular density into an accurate ground-state quantum mechanical density. We find that this model outperforms all prior density prediction approaches. Because the input is itself a real-space density, the predictions are equivariant to molecular symmetry transformations even though the model is not constructed to be. Due to its simplicity, the model is directly applicable to unseen molecular conformations and chemical elements. We show that fine-tuning on limited new data provides high accuracy even in challenging cases of exotic elements and charge states. Our work suggests new routes to learning real-space physical quantities drawing from the established ideas of image processing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_12335 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Image Super-resolution Inspired Electron Density Prediction Li, Chenghan Sharir, Or Yuan, Shunyue Chan, Garnet K. Chemical Physics Machine Learning Drawing inspiration from the domain of image super-resolution, we view the electron density as a 3D grayscale image and use a convolutional residual network to transform a crude and trivially generated guess of the molecular density into an accurate ground-state quantum mechanical density. We find that this model outperforms all prior density prediction approaches. Because the input is itself a real-space density, the predictions are equivariant to molecular symmetry transformations even though the model is not constructed to be. Due to its simplicity, the model is directly applicable to unseen molecular conformations and chemical elements. We show that fine-tuning on limited new data provides high accuracy even in challenging cases of exotic elements and charge states. Our work suggests new routes to learning real-space physical quantities drawing from the established ideas of image processing. |
| title | Image Super-resolution Inspired Electron Density Prediction |
| topic | Chemical Physics Machine Learning |
| url | https://arxiv.org/abs/2402.12335 |