Grand canonical generative diffusion model for crystalline phases and grain boundaries
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910580842954752 |
|---|---|
| author | Lei, Bo Chen, Enze Kwon, Hyuna Hsu, Tim Sadigh, Babak Lordi, Vincenzo Frolov, Timofey Zhou, Fei |
| author_facet | Lei, Bo Chen, Enze Kwon, Hyuna Hsu, Tim Sadigh, Babak Lordi, Vincenzo Frolov, Timofey Zhou, Fei |
| contents | The diffusion model has emerged as a powerful tool for generating atomic structures for materials science. This work calls attention to the deficiency of current particle-based diffusion models, which represent atoms as a point cloud, in generating even the simplest ordered crystalline structures. The problem is attributed to particles being trapped in local minima during the score-driven simulated annealing of the diffusion process, similar to the physical process of force-driven simulated annealing. We develop a solution, the grand canonical diffusion model, which adopts an alternative voxel-based representation with continuous rather than fixed number of particles. The method is applied towards generation of several common crystalline phases as well as the technologically important and challenging problem of grain boundary structures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_15601 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Grand canonical generative diffusion model for crystalline phases and grain boundaries Lei, Bo Chen, Enze Kwon, Hyuna Hsu, Tim Sadigh, Babak Lordi, Vincenzo Frolov, Timofey Zhou, Fei Materials Science Machine Learning The diffusion model has emerged as a powerful tool for generating atomic structures for materials science. This work calls attention to the deficiency of current particle-based diffusion models, which represent atoms as a point cloud, in generating even the simplest ordered crystalline structures. The problem is attributed to particles being trapped in local minima during the score-driven simulated annealing of the diffusion process, similar to the physical process of force-driven simulated annealing. We develop a solution, the grand canonical diffusion model, which adopts an alternative voxel-based representation with continuous rather than fixed number of particles. The method is applied towards generation of several common crystalline phases as well as the technologically important and challenging problem of grain boundary structures. |
| title | Grand canonical generative diffusion model for crystalline phases and grain boundaries |
| topic | Materials Science Machine Learning |
| url | https://arxiv.org/abs/2408.15601 |