Interpretable machine learning model of effective mass in perovskite oxides with cross-scale features
The interpretability of machine learning reveals associations between input features and predicted physical properties in models, which are essential for discovering new materials. However, previous works were mainly devoted to algorithm improvement, while the essential multi-scale characteristics a...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Journal of Materiomics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S235284782400042X |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841560561944363008 |
|---|---|
| author | Changjiao Li Zhengtao Huang Hua Hao Zhonghui Shen Guanghui Zhao Ben Xu Hanxing Liu |
| author_facet | Changjiao Li Zhengtao Huang Hua Hao Zhonghui Shen Guanghui Zhao Ben Xu Hanxing Liu |
| author_sort | Changjiao Li |
| collection | DOAJ |
| description | The interpretability of machine learning reveals associations between input features and predicted physical properties in models, which are essential for discovering new materials. However, previous works were mainly devoted to algorithm improvement, while the essential multi-scale characteristics are not well addressed. This paper introduces distortion modes of oxygen octahedrons as cross-scale structural features to bridge chemical compositions and material properties. Combining model-agnostic interpretation methods, we are able to achieve interpretability even using simple machine learning schemes and develop a predictive model of effective mass for a widely used material type, namely perovskite oxides. With this framework, we reach the interpretability of the model, understanding the trend of the effective mass without any prior background information. Moreover, we obtained the knowledge only available to experts, i.e., the interpretation of effective mass from the s–p orbitals hybridization of B-site cations and O2− in ABO3 perovskite oxides. |
| format | Article |
| id | doaj-art-7aec59c1830348658cd0f649a7563625 |
| institution | Kabale University |
| issn | 2352-8478 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Materiomics |
| spelling | doaj-art-7aec59c1830348658cd0f649a75636252025-01-04T04:56:34ZengElsevierJournal of Materiomics2352-84782025-01-01111100848Interpretable machine learning model of effective mass in perovskite oxides with cross-scale featuresChangjiao Li0Zhengtao Huang1Hua Hao2Zhonghui Shen3Guanghui Zhao4Ben Xu5Hanxing Liu6State Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Center for Smart Materials and Device Integration, School of Material Science and Engineering, Wuhan University of Technology, Wuhan, 430070, ChinaState Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Center for Smart Materials and Device Integration, School of Material Science and Engineering, Wuhan University of Technology, Wuhan, 430070, ChinaState Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Center for Smart Materials and Device Integration, School of Material Science and Engineering, Wuhan University of Technology, Wuhan, 430070, ChinaState Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Center for Smart Materials and Device Integration, School of Material Science and Engineering, Wuhan University of Technology, Wuhan, 430070, ChinaState Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Center for Smart Materials and Device Integration, School of Material Science and Engineering, Wuhan University of Technology, Wuhan, 430070, ChinaGraduate School of China Academy of Engineering Physics, Beijing, 100193, China; Corresponding author.State Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Center for Smart Materials and Device Integration, School of Material Science and Engineering, Wuhan University of Technology, Wuhan, 430070, China; Corresponding author.The interpretability of machine learning reveals associations between input features and predicted physical properties in models, which are essential for discovering new materials. However, previous works were mainly devoted to algorithm improvement, while the essential multi-scale characteristics are not well addressed. This paper introduces distortion modes of oxygen octahedrons as cross-scale structural features to bridge chemical compositions and material properties. Combining model-agnostic interpretation methods, we are able to achieve interpretability even using simple machine learning schemes and develop a predictive model of effective mass for a widely used material type, namely perovskite oxides. With this framework, we reach the interpretability of the model, understanding the trend of the effective mass without any prior background information. Moreover, we obtained the knowledge only available to experts, i.e., the interpretation of effective mass from the s–p orbitals hybridization of B-site cations and O2− in ABO3 perovskite oxides.http://www.sciencedirect.com/science/article/pii/S235284782400042XMachine learningPerovskite oxidesInterpretabilityEffective massCrystal structure |
| spellingShingle | Changjiao Li Zhengtao Huang Hua Hao Zhonghui Shen Guanghui Zhao Ben Xu Hanxing Liu Interpretable machine learning model of effective mass in perovskite oxides with cross-scale features Journal of Materiomics Machine learning Perovskite oxides Interpretability Effective mass Crystal structure |
| title | Interpretable machine learning model of effective mass in perovskite oxides with cross-scale features |
| title_full | Interpretable machine learning model of effective mass in perovskite oxides with cross-scale features |
| title_fullStr | Interpretable machine learning model of effective mass in perovskite oxides with cross-scale features |
| title_full_unstemmed | Interpretable machine learning model of effective mass in perovskite oxides with cross-scale features |
| title_short | Interpretable machine learning model of effective mass in perovskite oxides with cross-scale features |
| title_sort | interpretable machine learning model of effective mass in perovskite oxides with cross scale features |
| topic | Machine learning Perovskite oxides Interpretability Effective mass Crystal structure |
| url | http://www.sciencedirect.com/science/article/pii/S235284782400042X |
| work_keys_str_mv | AT changjiaoli interpretablemachinelearningmodelofeffectivemassinperovskiteoxideswithcrossscalefeatures AT zhengtaohuang interpretablemachinelearningmodelofeffectivemassinperovskiteoxideswithcrossscalefeatures AT huahao interpretablemachinelearningmodelofeffectivemassinperovskiteoxideswithcrossscalefeatures AT zhonghuishen interpretablemachinelearningmodelofeffectivemassinperovskiteoxideswithcrossscalefeatures AT guanghuizhao interpretablemachinelearningmodelofeffectivemassinperovskiteoxideswithcrossscalefeatures AT benxu interpretablemachinelearningmodelofeffectivemassinperovskiteoxideswithcrossscalefeatures AT hanxingliu interpretablemachinelearningmodelofeffectivemassinperovskiteoxideswithcrossscalefeatures |