Interpretable machine learning for stability and electronic structure prediction of Janus III–VI van der Waals heterostructures
Abstract Machine learning (ML) techniques have made enormous progress in the field of materials science. However, many conventional ML algorithms operate as “black‐boxes”, lacking transparency in revealing explicit relationships between material features and target properties. To address this, the d...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley-VCH
2024-12-01
|
| Series: | Materials Genome Engineering Advances |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/mgea.76 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841543276787662848 |
|---|---|
| author | Yudong Shi Yinggan Zhang Jiansen Wen Zhou Cui Jianhui Chen Xiaochun Huang Cuilian Wen Baisheng Sa Zhimei Sun |
| author_facet | Yudong Shi Yinggan Zhang Jiansen Wen Zhou Cui Jianhui Chen Xiaochun Huang Cuilian Wen Baisheng Sa Zhimei Sun |
| author_sort | Yudong Shi |
| collection | DOAJ |
| description | Abstract Machine learning (ML) techniques have made enormous progress in the field of materials science. However, many conventional ML algorithms operate as “black‐boxes”, lacking transparency in revealing explicit relationships between material features and target properties. To address this, the development of interpretable ML models is essential to drive further advancements in AI‐driven materials discovery. In this study, we present an interpretable framework that combines traditional machine learning with symbolic regression, using Janus III–VI vdW heterostructures as a case study. This approach enables fast and accurate predictions of stability and electronic structure. Our results demonstrate that the prediction accuracy using the classification model for stability, based on formation energy, reaches 0.960. On the other hand, the R2, MAE, and RMSE value using the regression model for electronic structure prediction, based on band gap, achieves 0.927, 0.113, and 0.141 on the testing set, respectively. Additionally, we identify a universal interpretable descriptor comprising five simple parameters that reveals the underlying physical relationships between the candidate heterostructures and their band gaps. This descriptor not only delivers high accuracy in band gap prediction but also provides explicit physical insight into the material properties. |
| format | Article |
| id | doaj-art-cfaeedb637ad4ff38d1db2bf05b4605a |
| institution | Kabale University |
| issn | 2940-9489 2940-9497 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Wiley-VCH |
| record_format | Article |
| series | Materials Genome Engineering Advances |
| spelling | doaj-art-cfaeedb637ad4ff38d1db2bf05b4605a2025-01-13T15:15:31ZengWiley-VCHMaterials Genome Engineering Advances2940-94892940-94972024-12-0124n/an/a10.1002/mgea.76Interpretable machine learning for stability and electronic structure prediction of Janus III–VI van der Waals heterostructuresYudong Shi0Yinggan Zhang1Jiansen Wen2Zhou Cui3Jianhui Chen4Xiaochun Huang5Cuilian Wen6Baisheng Sa7Zhimei Sun8Multiscale Computational Materials Facility & Materials Genome Institute School of Materials Science and Engineering Fuzhou University Fuzhou ChinaSchool of Materials Science and Engineering Beihang University Beijing ChinaMultiscale Computational Materials Facility & Materials Genome Institute School of Materials Science and Engineering Fuzhou University Fuzhou ChinaMultiscale Computational Materials Facility & Materials Genome Institute School of Materials Science and Engineering Fuzhou University Fuzhou ChinaMultiscale Computational Materials Facility & Materials Genome Institute School of Materials Science and Engineering Fuzhou University Fuzhou ChinaPhysikalisches institut, experimentelle Physik ii Julius‐Maximilians‐Universität Würzburg Würzburg GermanyMultiscale Computational Materials Facility & Materials Genome Institute School of Materials Science and Engineering Fuzhou University Fuzhou ChinaMultiscale Computational Materials Facility & Materials Genome Institute School of Materials Science and Engineering Fuzhou University Fuzhou ChinaSchool of Materials Science and Engineering Beihang University Beijing ChinaAbstract Machine learning (ML) techniques have made enormous progress in the field of materials science. However, many conventional ML algorithms operate as “black‐boxes”, lacking transparency in revealing explicit relationships between material features and target properties. To address this, the development of interpretable ML models is essential to drive further advancements in AI‐driven materials discovery. In this study, we present an interpretable framework that combines traditional machine learning with symbolic regression, using Janus III–VI vdW heterostructures as a case study. This approach enables fast and accurate predictions of stability and electronic structure. Our results demonstrate that the prediction accuracy using the classification model for stability, based on formation energy, reaches 0.960. On the other hand, the R2, MAE, and RMSE value using the regression model for electronic structure prediction, based on band gap, achieves 0.927, 0.113, and 0.141 on the testing set, respectively. Additionally, we identify a universal interpretable descriptor comprising five simple parameters that reveals the underlying physical relationships between the candidate heterostructures and their band gaps. This descriptor not only delivers high accuracy in band gap prediction but also provides explicit physical insight into the material properties.https://doi.org/10.1002/mgea.76descriptorinterpretable machine learningJanus III–VI van der Waals heterostructures |
| spellingShingle | Yudong Shi Yinggan Zhang Jiansen Wen Zhou Cui Jianhui Chen Xiaochun Huang Cuilian Wen Baisheng Sa Zhimei Sun Interpretable machine learning for stability and electronic structure prediction of Janus III–VI van der Waals heterostructures Materials Genome Engineering Advances descriptor interpretable machine learning Janus III–VI van der Waals heterostructures |
| title | Interpretable machine learning for stability and electronic structure prediction of Janus III–VI van der Waals heterostructures |
| title_full | Interpretable machine learning for stability and electronic structure prediction of Janus III–VI van der Waals heterostructures |
| title_fullStr | Interpretable machine learning for stability and electronic structure prediction of Janus III–VI van der Waals heterostructures |
| title_full_unstemmed | Interpretable machine learning for stability and electronic structure prediction of Janus III–VI van der Waals heterostructures |
| title_short | Interpretable machine learning for stability and electronic structure prediction of Janus III–VI van der Waals heterostructures |
| title_sort | interpretable machine learning for stability and electronic structure prediction of janus iii vi van der waals heterostructures |
| topic | descriptor interpretable machine learning Janus III–VI van der Waals heterostructures |
| url | https://doi.org/10.1002/mgea.76 |
| work_keys_str_mv | AT yudongshi interpretablemachinelearningforstabilityandelectronicstructurepredictionofjanusiiivivanderwaalsheterostructures AT yingganzhang interpretablemachinelearningforstabilityandelectronicstructurepredictionofjanusiiivivanderwaalsheterostructures AT jiansenwen interpretablemachinelearningforstabilityandelectronicstructurepredictionofjanusiiivivanderwaalsheterostructures AT zhoucui interpretablemachinelearningforstabilityandelectronicstructurepredictionofjanusiiivivanderwaalsheterostructures AT jianhuichen interpretablemachinelearningforstabilityandelectronicstructurepredictionofjanusiiivivanderwaalsheterostructures AT xiaochunhuang interpretablemachinelearningforstabilityandelectronicstructurepredictionofjanusiiivivanderwaalsheterostructures AT cuilianwen interpretablemachinelearningforstabilityandelectronicstructurepredictionofjanusiiivivanderwaalsheterostructures AT baishengsa interpretablemachinelearningforstabilityandelectronicstructurepredictionofjanusiiivivanderwaalsheterostructures AT zhimeisun interpretablemachinelearningforstabilityandelectronicstructurepredictionofjanusiiivivanderwaalsheterostructures |