Active learning accelerates ab initio molecular dynamics on reactive energy surfaces
© 2020 Elsevier Inc. Through autonomous data acquisition and machine learning, we demonstrate that our neural-network-based reactive force fields allow us to study the dynamical effects of several pericyclic reactions and to predict solvent effects on periselectivity. Our method is over 2,000 times...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Elsevier BV
2022
|
| Online Access: | https://hdl.handle.net/1721.1/142510 |
| _version_ | 1826205259464179712 |
|---|---|
| author | Ang, Shi Jun Wang, Wujie Schwalbe-Koda, Daniel Axelrod, Simon Gómez-Bombarelli, Rafael |
| author2 | Massachusetts Institute of Technology. Department of Materials Science and Engineering |
| author_facet | Massachusetts Institute of Technology. Department of Materials Science and Engineering Ang, Shi Jun Wang, Wujie Schwalbe-Koda, Daniel Axelrod, Simon Gómez-Bombarelli, Rafael |
| author_sort | Ang, Shi Jun |
| collection | MIT |
| description | © 2020 Elsevier Inc. Through autonomous data acquisition and machine learning, we demonstrate that our neural-network-based reactive force fields allow us to study the dynamical effects of several pericyclic reactions and to predict solvent effects on periselectivity. Our method is over 2,000 times faster than the traditional density functional theory approach, and its accuracy matches the parent quantum mechanical method. Given the efficiency of our machine learning framework, we envisage its applicability in studying larger reactive systems with a higher complexity. |
| first_indexed | 2024-09-23T13:09:53Z |
| format | Article |
| id | mit-1721.1/142510 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T13:09:53Z |
| publishDate | 2022 |
| publisher | Elsevier BV |
| record_format | dspace |
| spelling | mit-1721.1/1425102023-06-22T13:30:43Z Active learning accelerates ab initio molecular dynamics on reactive energy surfaces Ang, Shi Jun Wang, Wujie Schwalbe-Koda, Daniel Axelrod, Simon Gómez-Bombarelli, Rafael Massachusetts Institute of Technology. Department of Materials Science and Engineering © 2020 Elsevier Inc. Through autonomous data acquisition and machine learning, we demonstrate that our neural-network-based reactive force fields allow us to study the dynamical effects of several pericyclic reactions and to predict solvent effects on periselectivity. Our method is over 2,000 times faster than the traditional density functional theory approach, and its accuracy matches the parent quantum mechanical method. Given the efficiency of our machine learning framework, we envisage its applicability in studying larger reactive systems with a higher complexity. 2022-05-12T19:25:22Z 2022-05-12T19:25:22Z 2021 2022-05-12T19:21:05Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/142510 Ang, Shi Jun, Wang, Wujie, Schwalbe-Koda, Daniel, Axelrod, Simon and Gómez-Bombarelli, Rafael. 2021. "Active learning accelerates ab initio molecular dynamics on reactive energy surfaces." Chem, 7 (3). en 10.1016/J.CHEMPR.2020.12.009 Chem Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf Elsevier BV ChemRxiv |
| spellingShingle | Ang, Shi Jun Wang, Wujie Schwalbe-Koda, Daniel Axelrod, Simon Gómez-Bombarelli, Rafael Active learning accelerates ab initio molecular dynamics on reactive energy surfaces |
| title | Active learning accelerates ab initio molecular dynamics on reactive energy surfaces |
| title_full | Active learning accelerates ab initio molecular dynamics on reactive energy surfaces |
| title_fullStr | Active learning accelerates ab initio molecular dynamics on reactive energy surfaces |
| title_full_unstemmed | Active learning accelerates ab initio molecular dynamics on reactive energy surfaces |
| title_short | Active learning accelerates ab initio molecular dynamics on reactive energy surfaces |
| title_sort | active learning accelerates ab initio molecular dynamics on reactive energy surfaces |
| url | https://hdl.handle.net/1721.1/142510 |
| work_keys_str_mv | AT angshijun activelearningacceleratesabinitiomoleculardynamicsonreactiveenergysurfaces AT wangwujie activelearningacceleratesabinitiomoleculardynamicsonreactiveenergysurfaces AT schwalbekodadaniel activelearningacceleratesabinitiomoleculardynamicsonreactiveenergysurfaces AT axelrodsimon activelearningacceleratesabinitiomoleculardynamicsonreactiveenergysurfaces AT gomezbombarellirafael activelearningacceleratesabinitiomoleculardynamicsonreactiveenergysurfaces |