Learning to Model Atoms Across Scales
The understanding of atoms and how they interact forms the foundation of modern natural science, as well as material and drug discovery efforts. Computational chemistry methods such as density functional theory and molecular dynamics simulation can offer an unparalleled spatiotemporal resolution for...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Massachusetts Institute of Technology
2024
|
| Online Access: | https://hdl.handle.net/1721.1/156328 |
| _version_ | 1826196791525113856 |
|---|---|
| author | Fu, Xiang |
| author2 | Jaakkola, Tommi S. |
| author_facet | Jaakkola, Tommi S. Fu, Xiang |
| author_sort | Fu, Xiang |
| collection | MIT |
| description | The understanding of atoms and how they interact forms the foundation of modern natural science, as well as material and drug discovery efforts. Computational chemistry methods such as density functional theory and molecular dynamics simulation can offer an unparalleled spatiotemporal resolution for observing microscopic mechanisms and predicting macroscopic phenomena. However, many natural processes are extremely complex, requiring highly accurate modeling of many atoms for a considerable period to study. Computational chemistry methods may not be accurate or efficient enough, limiting the applicable domains and scales. Furthermore, discovering new materials and drugs requires novel candidate atomistic structures, which are conventionally based on heuristic or exhaustive search methods. This thesis presents machine learning methods for modeling atoms for tasks across different scales. First, we propose machine learning force fields that can decompose molecular interactions into fast and slow components, and then accelerate molecular simulations through multiscale integration. Second, we propose an end-to-end workflow for learning time-integrated coarse-grained molecular dynamics using multi-scale graph neural networks. Third, we propose diffusion models designed for periodic material structures that can enable the discovery of novel stable materials as well as material inverse design given a target property. The material diffusion model can be further extended to complex metal-organic frameworks with a multi-scale modeling approach. |
| first_indexed | 2024-09-23T10:37:43Z |
| format | Thesis |
| id | mit-1721.1/156328 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T10:37:43Z |
| publishDate | 2024 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1563282024-08-22T03:46:11Z Learning to Model Atoms Across Scales Fu, Xiang Jaakkola, Tommi S. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science The understanding of atoms and how they interact forms the foundation of modern natural science, as well as material and drug discovery efforts. Computational chemistry methods such as density functional theory and molecular dynamics simulation can offer an unparalleled spatiotemporal resolution for observing microscopic mechanisms and predicting macroscopic phenomena. However, many natural processes are extremely complex, requiring highly accurate modeling of many atoms for a considerable period to study. Computational chemistry methods may not be accurate or efficient enough, limiting the applicable domains and scales. Furthermore, discovering new materials and drugs requires novel candidate atomistic structures, which are conventionally based on heuristic or exhaustive search methods. This thesis presents machine learning methods for modeling atoms for tasks across different scales. First, we propose machine learning force fields that can decompose molecular interactions into fast and slow components, and then accelerate molecular simulations through multiscale integration. Second, we propose an end-to-end workflow for learning time-integrated coarse-grained molecular dynamics using multi-scale graph neural networks. Third, we propose diffusion models designed for periodic material structures that can enable the discovery of novel stable materials as well as material inverse design given a target property. The material diffusion model can be further extended to complex metal-organic frameworks with a multi-scale modeling approach. Ph.D. 2024-08-21T18:57:05Z 2024-08-21T18:57:05Z 2024-05 2024-07-10T13:01:34.538Z Thesis https://hdl.handle.net/1721.1/156328 Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) Copyright retained by author(s) https://creativecommons.org/licenses/by-nc-nd/4.0/ application/pdf Massachusetts Institute of Technology |
| spellingShingle | Fu, Xiang Learning to Model Atoms Across Scales |
| title | Learning to Model Atoms Across Scales |
| title_full | Learning to Model Atoms Across Scales |
| title_fullStr | Learning to Model Atoms Across Scales |
| title_full_unstemmed | Learning to Model Atoms Across Scales |
| title_short | Learning to Model Atoms Across Scales |
| title_sort | learning to model atoms across scales |
| url | https://hdl.handle.net/1721.1/156328 |
| work_keys_str_mv | AT fuxiang learningtomodelatomsacrossscales |