Machine learning configuration-dependent friction tensors in Langevin heatbaths
Dynamics of coarse-grained particle systems derived via the Mori–Zwanzig projection formalism commonly take the form of a (generalized) Langevin equation with configuration-dependent friction tensor and diffusion coefficient matrix. In this article, we introduce a class of equivariant representation...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2025-01-01
|
| Series: | Machine Learning: Science and Technology |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/2632-2153/ada248 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583620771971072 |
|---|---|
| author | Matthias Sachs Wojciech G Stark Reinhard J Maurer Christoph Ortner |
| author_facet | Matthias Sachs Wojciech G Stark Reinhard J Maurer Christoph Ortner |
| author_sort | Matthias Sachs |
| collection | DOAJ |
| description | Dynamics of coarse-grained particle systems derived via the Mori–Zwanzig projection formalism commonly take the form of a (generalized) Langevin equation with configuration-dependent friction tensor and diffusion coefficient matrix. In this article, we introduce a class of equivariant representations of tensor-valued functions based on the Atomic Cluster Expansion framework that allows for efficient learning of such configuration-dependent friction tensors from data. Besides satisfying the correct equivariance properties with respect to the Euclidean group E(3), the resulting heat bath models satisfy a fluctuation-dissipation relation. We demonstrate the capabilities of the model approach by fitting a model of configuration-dependent tensorial electronic friction calculated from first principles that arises during reactive molecular dynamics at metal surfaces. |
| format | Article |
| id | doaj-art-cebc92c22f3e43f79c159c13ba9f0296 |
| institution | Kabale University |
| issn | 2632-2153 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Machine Learning: Science and Technology |
| spelling | doaj-art-cebc92c22f3e43f79c159c13ba9f02962025-01-28T10:10:36ZengIOP PublishingMachine Learning: Science and Technology2632-21532025-01-016101501610.1088/2632-2153/ada248Machine learning configuration-dependent friction tensors in Langevin heatbathsMatthias Sachs0https://orcid.org/0000-0002-9003-337XWojciech G Stark1https://orcid.org/0000-0001-6279-2638Reinhard J Maurer2https://orcid.org/0000-0002-3004-785XChristoph Ortner3https://orcid.org/0000-0003-1498-8120School of Mathematics, University of Birmingham , Ring Rd N, Birmingham B15 2TS, United KingdomDepartment of Chemistry, University of Warwick , Gibbet Hill Rd, Coventry CV4 7AL, United KingdomDepartment of Chemistry, University of Warwick , Gibbet Hill Rd, Coventry CV4 7AL, United Kingdom; Department of Physics, University of Warwick , Gibbet Hill Rd, Coventry CV4 7AL, United KingdomDepartment of Mathematics, University of British Columbia , Vancouver, BC V6T 1Z2, CanadaDynamics of coarse-grained particle systems derived via the Mori–Zwanzig projection formalism commonly take the form of a (generalized) Langevin equation with configuration-dependent friction tensor and diffusion coefficient matrix. In this article, we introduce a class of equivariant representations of tensor-valued functions based on the Atomic Cluster Expansion framework that allows for efficient learning of such configuration-dependent friction tensors from data. Besides satisfying the correct equivariance properties with respect to the Euclidean group E(3), the resulting heat bath models satisfy a fluctuation-dissipation relation. We demonstrate the capabilities of the model approach by fitting a model of configuration-dependent tensorial electronic friction calculated from first principles that arises during reactive molecular dynamics at metal surfaces.https://doi.org/10.1088/2632-2153/ada248atomic cluster expansiondensity functional theorydynamics at metal surfaceselectronic friction tensorLangevin equationequivariant representation |
| spellingShingle | Matthias Sachs Wojciech G Stark Reinhard J Maurer Christoph Ortner Machine learning configuration-dependent friction tensors in Langevin heatbaths Machine Learning: Science and Technology atomic cluster expansion density functional theory dynamics at metal surfaces electronic friction tensor Langevin equation equivariant representation |
| title | Machine learning configuration-dependent friction tensors in Langevin heatbaths |
| title_full | Machine learning configuration-dependent friction tensors in Langevin heatbaths |
| title_fullStr | Machine learning configuration-dependent friction tensors in Langevin heatbaths |
| title_full_unstemmed | Machine learning configuration-dependent friction tensors in Langevin heatbaths |
| title_short | Machine learning configuration-dependent friction tensors in Langevin heatbaths |
| title_sort | machine learning configuration dependent friction tensors in langevin heatbaths |
| topic | atomic cluster expansion density functional theory dynamics at metal surfaces electronic friction tensor Langevin equation equivariant representation |
| url | https://doi.org/10.1088/2632-2153/ada248 |
| work_keys_str_mv | AT matthiassachs machinelearningconfigurationdependentfrictiontensorsinlangevinheatbaths AT wojciechgstark machinelearningconfigurationdependentfrictiontensorsinlangevinheatbaths AT reinhardjmaurer machinelearningconfigurationdependentfrictiontensorsinlangevinheatbaths AT christophortner machinelearningconfigurationdependentfrictiontensorsinlangevinheatbaths |