A multi-objective, multi-interpretable machine learning demonstration verified by domain knowledge for ductile thermoelectric materials
Multi-objective machine learning (ML) methods are widely used in the field of materials because material optimizations are always multi-objective. Traditional multi-objective optimization methods mainly use a combination of hierarchical single-objective optimization. However, this strategy often has...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Journal of Materiomics |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352847824001126 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841542331772174336 |
|---|---|
| author | Xiangdong Wang Yan Cao Jialin Ji Ye Sheng Jiong Yang Xuezhi Ke |
| author_facet | Xiangdong Wang Yan Cao Jialin Ji Ye Sheng Jiong Yang Xuezhi Ke |
| author_sort | Xiangdong Wang |
| collection | DOAJ |
| description | Multi-objective machine learning (ML) methods are widely used in the field of materials because material optimizations are always multi-objective. Traditional multi-objective optimization methods mainly use a combination of hierarchical single-objective optimization. However, this strategy often has difficulty in finding features that can optimize multiple objectives simultaneously. In this work, taking the two objectives of ductility and thermoelectric performance as examples, interpretable and explainable ML strategies are used to find features that can simultaneously optimize multiple objectives. Specifically, SHAP and SISSO are applied for qualitative analysis and quantitative analysis between key features and target values. Both SISSO and SHAP show that EN(ab)A/B and V are both positively correlated with zT and negatively correlated with Pugh's ratio. Furthermore, domain knowledge helps to rationalize the two favorable features. The compounds with large EN(ab)A/B tend to have high band degeneracies, resulting in high zT. High EN(ab)A/B correspond to weak B–X bonds, reducing the G and Pugh's ratio, and improving the ductility of materials. On the other hand, large V will cause small G, which is beneficial to small Pugh's ratio and large zT (via low κL). The present work demonstrates the significance of multi-objective optimization and domain knowledge in the development of materials informatics. |
| format | Article |
| id | doaj-art-41bd1e5a94ec446eafd77219b9634ee5 |
| institution | Kabale University |
| issn | 2352-8478 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Materiomics |
| spelling | doaj-art-41bd1e5a94ec446eafd77219b9634ee52025-01-14T04:12:30ZengElsevierJournal of Materiomics2352-84782025-03-01112100886A multi-objective, multi-interpretable machine learning demonstration verified by domain knowledge for ductile thermoelectric materialsXiangdong Wang0Yan Cao1Jialin Ji2Ye Sheng3Jiong Yang4Xuezhi Ke5School of Physics and Electronic Science, East China Normal University, Shanghai, 200241, ChinaDepartment of Architecture, College of Civil Engineering and Architecture, Zhejiang University, Hangzhou, 310027, Zhejiang, ChinaCollege of Biological, Chemical Sciences and Engineering, Jiaxing University, Jiaxing, 314001, Zhejiang, ChinaDepartment of Materials Science and Engineering, Southern University of Science and Technology, Shenzhen, 518055, Guangdong, China; Corresponding author.Materials Genome Institute, Shanghai University, Shanghai, 200444, ChinaSchool of Physics and Electronic Science, East China Normal University, Shanghai, 200241, China; Corresponding author.Multi-objective machine learning (ML) methods are widely used in the field of materials because material optimizations are always multi-objective. Traditional multi-objective optimization methods mainly use a combination of hierarchical single-objective optimization. However, this strategy often has difficulty in finding features that can optimize multiple objectives simultaneously. In this work, taking the two objectives of ductility and thermoelectric performance as examples, interpretable and explainable ML strategies are used to find features that can simultaneously optimize multiple objectives. Specifically, SHAP and SISSO are applied for qualitative analysis and quantitative analysis between key features and target values. Both SISSO and SHAP show that EN(ab)A/B and V are both positively correlated with zT and negatively correlated with Pugh's ratio. Furthermore, domain knowledge helps to rationalize the two favorable features. The compounds with large EN(ab)A/B tend to have high band degeneracies, resulting in high zT. High EN(ab)A/B correspond to weak B–X bonds, reducing the G and Pugh's ratio, and improving the ductility of materials. On the other hand, large V will cause small G, which is beneficial to small Pugh's ratio and large zT (via low κL). The present work demonstrates the significance of multi-objective optimization and domain knowledge in the development of materials informatics.http://www.sciencedirect.com/science/article/pii/S2352847824001126multi-objectiveductile thermoelectric materialsmulti-interpretable machine learning |
| spellingShingle | Xiangdong Wang Yan Cao Jialin Ji Ye Sheng Jiong Yang Xuezhi Ke A multi-objective, multi-interpretable machine learning demonstration verified by domain knowledge for ductile thermoelectric materials Journal of Materiomics multi-objective ductile thermoelectric materials multi-interpretable machine learning |
| title | A multi-objective, multi-interpretable machine learning demonstration verified by domain knowledge for ductile thermoelectric materials |
| title_full | A multi-objective, multi-interpretable machine learning demonstration verified by domain knowledge for ductile thermoelectric materials |
| title_fullStr | A multi-objective, multi-interpretable machine learning demonstration verified by domain knowledge for ductile thermoelectric materials |
| title_full_unstemmed | A multi-objective, multi-interpretable machine learning demonstration verified by domain knowledge for ductile thermoelectric materials |
| title_short | A multi-objective, multi-interpretable machine learning demonstration verified by domain knowledge for ductile thermoelectric materials |
| title_sort | multi objective multi interpretable machine learning demonstration verified by domain knowledge for ductile thermoelectric materials |
| topic | multi-objective ductile thermoelectric materials multi-interpretable machine learning |
| url | http://www.sciencedirect.com/science/article/pii/S2352847824001126 |
| work_keys_str_mv | AT xiangdongwang amultiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT yancao amultiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT jialinji amultiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT yesheng amultiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT jiongyang amultiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT xuezhike amultiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT xiangdongwang multiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT yancao multiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT jialinji multiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT yesheng multiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT jiongyang multiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials AT xuezhike multiobjectivemultiinterpretablemachinelearningdemonstrationverifiedbydomainknowledgeforductilethermoelectricmaterials |