Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2509.24115 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914062583988224 |
|---|---|
| author | Dramko, Evan Xiong, Yihuang Zhu, Yizhi Hautier, Geoffroy Reps, Thomas Jermaine, Christopher Kyrillidis, Anastasios |
| author_facet | Dramko, Evan Xiong, Yihuang Zhu, Yizhi Hautier, Geoffroy Reps, Thomas Jermaine, Christopher Kyrillidis, Anastasios |
| contents | Point defects play a central role in driving the properties of materials. First-principles methods are widely used to compute defect energetics and structures, including at scale for high-throughput defect databases. However, these methods are computationally expensive, making machine-learning force fields (MLFFs) an attractive alternative for accelerating structural relaxations. Most existing MLFFs are based on graph neural networks (GNNs), which can suffer from oversmoothing and poor representation of long-range interactions. Both of these issues are especially of concern when modeling point defects. To address these challenges, we introduce the Accelerated Deep Atomic Potential Transformer (ADAPT), an MLFF that replaces graph representations with a direct coordinates-in-space formulation and explicitly considers all pairwise atomic interactions. Atoms are treated as tokens, with a Transformer encoder modeling their interactions. Applied to a dataset of silicon point defects, ADAPT achieves a roughly 33 percent reduction in both force and energy prediction errors relative to a state-of-the-art GNN-based model, while requiring only a fraction of the computational cost. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_24115 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ADAPT: Lightweight, Long-Range Machine Learning Force Fields Without Graphs Dramko, Evan Xiong, Yihuang Zhu, Yizhi Hautier, Geoffroy Reps, Thomas Jermaine, Christopher Kyrillidis, Anastasios Machine Learning Materials Science Optimization and Control 68Q32 (Primary), 68T07 (Secondary) Point defects play a central role in driving the properties of materials. First-principles methods are widely used to compute defect energetics and structures, including at scale for high-throughput defect databases. However, these methods are computationally expensive, making machine-learning force fields (MLFFs) an attractive alternative for accelerating structural relaxations. Most existing MLFFs are based on graph neural networks (GNNs), which can suffer from oversmoothing and poor representation of long-range interactions. Both of these issues are especially of concern when modeling point defects. To address these challenges, we introduce the Accelerated Deep Atomic Potential Transformer (ADAPT), an MLFF that replaces graph representations with a direct coordinates-in-space formulation and explicitly considers all pairwise atomic interactions. Atoms are treated as tokens, with a Transformer encoder modeling their interactions. Applied to a dataset of silicon point defects, ADAPT achieves a roughly 33 percent reduction in both force and energy prediction errors relative to a state-of-the-art GNN-based model, while requiring only a fraction of the computational cost. |
| title | ADAPT: Lightweight, Long-Range Machine Learning Force Fields Without Graphs |
| topic | Machine Learning Materials Science Optimization and Control 68Q32 (Primary), 68T07 (Secondary) |
| url | https://arxiv.org/abs/2509.24115 |