Hybrid Ensembles for Improved Force Matching
Force matching is a method for parameterizing empirical potentials in which the empirical parameters are fitted to a reference potential energy surface (PES). Typically, training data are sampled from a canonical ensemble generated with either the empirical potential or the reference PES. In this Co...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
American Institute of Physics
2012
|
| Online Access: | http://hdl.handle.net/1721.1/69604 https://orcid.org/0000-0001-7111-0176 |
| _version_ | 1826212302413627392 |
|---|---|
| author | Wang, Lee-Ping Van Voorhis, Troy |
| author2 | Massachusetts Institute of Technology. Department of Chemistry |
| author_facet | Massachusetts Institute of Technology. Department of Chemistry Wang, Lee-Ping Van Voorhis, Troy |
| author_sort | Wang, Lee-Ping |
| collection | MIT |
| description | Force matching is a method for parameterizing empirical potentials in which the empirical parameters are fitted to a reference potential energy surface (PES). Typically, training data are sampled from a canonical ensemble generated with either the empirical potential or the reference PES. In this Communication, we show that sampling from either ensemble risks excluding critical regions of configuration space, leading to fitted potentials that deviate significantly from the reference PES. We present a hybrid ensemble which combines the Boltzmann probabilities of both potential surfaces into the fitting procedure, and we demonstrate that this technique improves the quality and stability of empirical potentials. |
| first_indexed | 2024-09-23T15:19:33Z |
| format | Article |
| id | mit-1721.1/69604 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T15:19:33Z |
| publishDate | 2012 |
| publisher | American Institute of Physics |
| record_format | dspace |
| spelling | mit-1721.1/696042022-09-29T14:12:18Z Hybrid Ensembles for Improved Force Matching Communication: Hybrid ensembles for improved force matching Wang, Lee-Ping Van Voorhis, Troy Massachusetts Institute of Technology. Department of Chemistry Van Voorhis, Troy Van Voorhis, Troy Wang, Lee-Ping Force matching is a method for parameterizing empirical potentials in which the empirical parameters are fitted to a reference potential energy surface (PES). Typically, training data are sampled from a canonical ensemble generated with either the empirical potential or the reference PES. In this Communication, we show that sampling from either ensemble risks excluding critical regions of configuration space, leading to fitted potentials that deviate significantly from the reference PES. We present a hybrid ensemble which combines the Boltzmann probabilities of both potential surfaces into the fitting procedure, and we demonstrate that this technique improves the quality and stability of empirical potentials. Eni S.p.A. (Firm) (Solar Frontiers Research Program) 2012-03-08T19:25:07Z 2012-03-08T19:25:07Z 2010-12 2010-09 Article http://purl.org/eprint/type/JournalArticle 0021-9606 1089-7690 http://hdl.handle.net/1721.1/69604 Wang, Lee-Ping, and Troy Van Voorhis. “Communication: Hybrid Ensembles for Improved Force Matching.” The Journal of Chemical Physics 133.23 (2010): 231101. https://orcid.org/0000-0001-7111-0176 en_US http://dx.doi.org/10.1063/1.3519043 Journal of Chemical Physics Creative Commons Attribution-Noncommercial-Share Alike 3.0 http://creativecommons.org/licenses/by-nc-sa/3.0/ application/pdf American Institute of Physics Prof. Van Voorhis via Erja Kajosalo |
| spellingShingle | Wang, Lee-Ping Van Voorhis, Troy Hybrid Ensembles for Improved Force Matching |
| title | Hybrid Ensembles for Improved Force Matching |
| title_full | Hybrid Ensembles for Improved Force Matching |
| title_fullStr | Hybrid Ensembles for Improved Force Matching |
| title_full_unstemmed | Hybrid Ensembles for Improved Force Matching |
| title_short | Hybrid Ensembles for Improved Force Matching |
| title_sort | hybrid ensembles for improved force matching |
| url | http://hdl.handle.net/1721.1/69604 https://orcid.org/0000-0001-7111-0176 |
| work_keys_str_mv | AT wangleeping hybridensemblesforimprovedforcematching AT vanvoorhistroy hybridensemblesforimprovedforcematching AT wangleeping communicationhybridensemblesforimprovedforcematching AT vanvoorhistroy communicationhybridensemblesforimprovedforcematching |