Computing formation enthalpies through an explainable machine learning method: the case of lanthanide orthophosphates solid solutions
In the last decade, the use of AI in Condensed Matter physics has seen a steep increase in the number of problems tackled and methods employed. A number of distinct Machine Learning approaches have been employed in many different topics; from prediction of material properties to computation of Densi...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-04-01
|
| Series: | Frontiers in Applied Mathematics and Statistics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fams.2024.1355726/full |
| _version_ | 1797230505696952320 |
|---|---|
| author | Edoardo Di Napoli Edoardo Di Napoli Xinzhe Wu Thomas Bornhake Piotr M. Kowalski Piotr M. Kowalski Piotr M. Kowalski |
| author_facet | Edoardo Di Napoli Edoardo Di Napoli Xinzhe Wu Thomas Bornhake Piotr M. Kowalski Piotr M. Kowalski Piotr M. Kowalski |
| author_sort | Edoardo Di Napoli |
| collection | DOAJ |
| description | In the last decade, the use of AI in Condensed Matter physics has seen a steep increase in the number of problems tackled and methods employed. A number of distinct Machine Learning approaches have been employed in many different topics; from prediction of material properties to computation of Density Functional Theory potentials and inter-atomic force fields. In many cases, the result is a surrogate model which returns promising predictions but is opaque on the inner mechanisms of its success. On the other hand, the typical practitioner looks for answers that are explainable and provide a clear insight into the mechanisms governing a physical phenomena. In this study, we describe a proposal to use a sophisticated combination of traditional Machine Learning methods to obtain an explainable model that outputs an explicit functional formulation for the material property of interest. We demonstrate the effectiveness of our methodology in deriving a new highly accurate expression for the enthalpy of formation of solid solutions of lanthanide orthophosphates. |
| first_indexed | 2024-04-24T15:29:33Z |
| format | Article |
| id | doaj.art-1a5d0b6648bd42a0b54cd7cad001e034 |
| institution | Directory Open Access Journal |
| issn | 2297-4687 |
| language | English |
| last_indexed | 2024-04-24T15:29:33Z |
| publishDate | 2024-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Applied Mathematics and Statistics |
| spelling | doaj.art-1a5d0b6648bd42a0b54cd7cad001e0342024-04-02T04:45:29ZengFrontiers Media S.A.Frontiers in Applied Mathematics and Statistics2297-46872024-04-011010.3389/fams.2024.13557261355726Computing formation enthalpies through an explainable machine learning method: the case of lanthanide orthophosphates solid solutionsEdoardo Di Napoli0Edoardo Di Napoli1Xinzhe Wu2Thomas Bornhake3Piotr M. Kowalski4Piotr M. Kowalski5Piotr M. Kowalski6Simulation and Data Lab Quantum Materials, Jülich Supercomputing Centre, Forschungszentrum Jülich, Jülich, GermanyJülich Aachen Research Alliance Center for Simulation and Data Science, Forschungszentrum Jülich, Jülich, GermanySimulation and Data Lab Quantum Materials, Jülich Supercomputing Centre, Forschungszentrum Jülich, Jülich, GermanyPhysics Department, RWTH Aachen University, Aachen, GermanyJülich Aachen Research Alliance Center for Simulation and Data Science, Forschungszentrum Jülich, Jülich, GermanyInstitute of Energy and Climate Research (IEK-13), Forschungszentrum Jülich, Jülich, GermanyJülich Aachen Research Alliance Energy, Forschungszentrum Jülich, Jülich, GermanyIn the last decade, the use of AI in Condensed Matter physics has seen a steep increase in the number of problems tackled and methods employed. A number of distinct Machine Learning approaches have been employed in many different topics; from prediction of material properties to computation of Density Functional Theory potentials and inter-atomic force fields. In many cases, the result is a surrogate model which returns promising predictions but is opaque on the inner mechanisms of its success. On the other hand, the typical practitioner looks for answers that are explainable and provide a clear insight into the mechanisms governing a physical phenomena. In this study, we describe a proposal to use a sophisticated combination of traditional Machine Learning methods to obtain an explainable model that outputs an explicit functional formulation for the material property of interest. We demonstrate the effectiveness of our methodology in deriving a new highly accurate expression for the enthalpy of formation of solid solutions of lanthanide orthophosphates.https://www.frontiersin.org/articles/10.3389/fams.2024.1355726/fullexplainable learningenthalpyLASSOridge regressionsparsificationsolid solutions |
| spellingShingle | Edoardo Di Napoli Edoardo Di Napoli Xinzhe Wu Thomas Bornhake Piotr M. Kowalski Piotr M. Kowalski Piotr M. Kowalski Computing formation enthalpies through an explainable machine learning method: the case of lanthanide orthophosphates solid solutions Frontiers in Applied Mathematics and Statistics explainable learning enthalpy LASSO ridge regression sparsification solid solutions |
| title | Computing formation enthalpies through an explainable machine learning method: the case of lanthanide orthophosphates solid solutions |
| title_full | Computing formation enthalpies through an explainable machine learning method: the case of lanthanide orthophosphates solid solutions |
| title_fullStr | Computing formation enthalpies through an explainable machine learning method: the case of lanthanide orthophosphates solid solutions |
| title_full_unstemmed | Computing formation enthalpies through an explainable machine learning method: the case of lanthanide orthophosphates solid solutions |
| title_short | Computing formation enthalpies through an explainable machine learning method: the case of lanthanide orthophosphates solid solutions |
| title_sort | computing formation enthalpies through an explainable machine learning method the case of lanthanide orthophosphates solid solutions |
| topic | explainable learning enthalpy LASSO ridge regression sparsification solid solutions |
| url | https://www.frontiersin.org/articles/10.3389/fams.2024.1355726/full |
| work_keys_str_mv | AT edoardodinapoli computingformationenthalpiesthroughanexplainablemachinelearningmethodthecaseoflanthanideorthophosphatessolidsolutions AT edoardodinapoli computingformationenthalpiesthroughanexplainablemachinelearningmethodthecaseoflanthanideorthophosphatessolidsolutions AT xinzhewu computingformationenthalpiesthroughanexplainablemachinelearningmethodthecaseoflanthanideorthophosphatessolidsolutions AT thomasbornhake computingformationenthalpiesthroughanexplainablemachinelearningmethodthecaseoflanthanideorthophosphatessolidsolutions AT piotrmkowalski computingformationenthalpiesthroughanexplainablemachinelearningmethodthecaseoflanthanideorthophosphatessolidsolutions AT piotrmkowalski computingformationenthalpiesthroughanexplainablemachinelearningmethodthecaseoflanthanideorthophosphatessolidsolutions AT piotrmkowalski computingformationenthalpiesthroughanexplainablemachinelearningmethodthecaseoflanthanideorthophosphatessolidsolutions |