Refining potential energy surface through dynamical properties via differentiable molecular simulation
Abstract Recently, machine learning potential (MLP) largely enhances the reliability of molecular dynamics, but its accuracy is limited by the underlying ab initio methods. A viable approach to overcome this limitation is to refine the potential by learning from experimental data, which now can be d...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-025-56061-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594657927757824 |
|---|---|
| author | Bin Han Kuang Yu |
| author_facet | Bin Han Kuang Yu |
| author_sort | Bin Han |
| collection | DOAJ |
| description | Abstract Recently, machine learning potential (MLP) largely enhances the reliability of molecular dynamics, but its accuracy is limited by the underlying ab initio methods. A viable approach to overcome this limitation is to refine the potential by learning from experimental data, which now can be done efficiently using modern automatic differentiation technique. However, potential refinement is mostly performed using thermodynamic properties, leaving the most accessible and informative dynamical data (like spectroscopy) unexploited. In this work, through a comprehensive application of adjoint and gradient truncation methods, we show that both memory and gradient explosion issues can be circumvented in many situations, so the dynamical property differentiation is well-behaved. Consequently, both transport coefficients and spectroscopic data can be used to improve the density functional theory based MLP towards higher accuracy. Essentially, this work contributes to the solution of the inverse problem of spectroscopy by extracting microscopic interactions from vibrational spectroscopic data. |
| format | Article |
| id | doaj-art-e462705becd1420e98b80ca88ce84ca6 |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-e462705becd1420e98b80ca88ce84ca62025-01-19T12:29:44ZengNature PortfolioNature Communications2041-17232025-01-0116111210.1038/s41467-025-56061-zRefining potential energy surface through dynamical properties via differentiable molecular simulationBin Han0Kuang Yu1Institute of Materials Research, Tsinghua Shenzhen International Graduate School (TSIGS)Institute of Materials Research, Tsinghua Shenzhen International Graduate School (TSIGS)Abstract Recently, machine learning potential (MLP) largely enhances the reliability of molecular dynamics, but its accuracy is limited by the underlying ab initio methods. A viable approach to overcome this limitation is to refine the potential by learning from experimental data, which now can be done efficiently using modern automatic differentiation technique. However, potential refinement is mostly performed using thermodynamic properties, leaving the most accessible and informative dynamical data (like spectroscopy) unexploited. In this work, through a comprehensive application of adjoint and gradient truncation methods, we show that both memory and gradient explosion issues can be circumvented in many situations, so the dynamical property differentiation is well-behaved. Consequently, both transport coefficients and spectroscopic data can be used to improve the density functional theory based MLP towards higher accuracy. Essentially, this work contributes to the solution of the inverse problem of spectroscopy by extracting microscopic interactions from vibrational spectroscopic data.https://doi.org/10.1038/s41467-025-56061-z |
| spellingShingle | Bin Han Kuang Yu Refining potential energy surface through dynamical properties via differentiable molecular simulation Nature Communications |
| title | Refining potential energy surface through dynamical properties via differentiable molecular simulation |
| title_full | Refining potential energy surface through dynamical properties via differentiable molecular simulation |
| title_fullStr | Refining potential energy surface through dynamical properties via differentiable molecular simulation |
| title_full_unstemmed | Refining potential energy surface through dynamical properties via differentiable molecular simulation |
| title_short | Refining potential energy surface through dynamical properties via differentiable molecular simulation |
| title_sort | refining potential energy surface through dynamical properties via differentiable molecular simulation |
| url | https://doi.org/10.1038/s41467-025-56061-z |
| work_keys_str_mv | AT binhan refiningpotentialenergysurfacethroughdynamicalpropertiesviadifferentiablemolecularsimulation AT kuangyu refiningpotentialenergysurfacethroughdynamicalpropertiesviadifferentiablemolecularsimulation |