Accelerating materials property prediction via a hybrid Transformer Graph framework that leverages four body interactions
Abstract Machine learning has advanced the rapid prediction of inorganic materials properties, yet data scarcity for specific properties and capturing thermodynamic stability remains challenging. We propose a framework utilizing a Graph Neural Network with composition-based and crystal structure-bas...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | npj Computational Materials |
| Online Access: | https://doi.org/10.1038/s41524-024-01472-7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594567923236864 |
|---|---|
| author | Mohammad Madani Valentina Lacivita Yongwoo Shin Anna Tarakanova |
| author_facet | Mohammad Madani Valentina Lacivita Yongwoo Shin Anna Tarakanova |
| author_sort | Mohammad Madani |
| collection | DOAJ |
| description | Abstract Machine learning has advanced the rapid prediction of inorganic materials properties, yet data scarcity for specific properties and capturing thermodynamic stability remains challenging. We propose a framework utilizing a Graph Neural Network with composition-based and crystal structure-based architectures, combined with a transfer learning scheme. This approach accurately predicts energy-related properties (e.g., total energy, energy above the convex hull, energy band gap) and data-scarce mechanical properties (e.g., bulk and shear modulus). Our model incorporates four-body interactions, capturing periodicity and structural characteristics. It outperforms state-of-the-art models in 8 materials property regression tasks. Also, this model predicts local atomic environments and global structural features better than several models. Transfer learning addresses mechanical property data scarcity, while separate architecture analysis allows application to materials lacking crystal structure information. Our framework’s interpretability aids in understanding elemental contributions, enhancing material design and discovery. Continuous advancements promise further performance improvements, driving efficient and accurate materials property prediction. |
| format | Article |
| id | doaj-art-b52920e6135242b2b774f66dfd76758f |
| institution | Kabale University |
| issn | 2057-3960 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Computational Materials |
| spelling | doaj-art-b52920e6135242b2b774f66dfd76758f2025-01-19T12:32:29ZengNature Portfolionpj Computational Materials2057-39602025-01-0111111410.1038/s41524-024-01472-7Accelerating materials property prediction via a hybrid Transformer Graph framework that leverages four body interactionsMohammad Madani0Valentina Lacivita1Yongwoo Shin2Anna Tarakanova3School of Mechanical, Aerospace, and Manufacturing Engineering, University of ConnecticutAdvanced Materials Lab, Samsung Semiconductor IncAdvanced Materials Lab, Samsung Semiconductor IncSchool of Mechanical, Aerospace, and Manufacturing Engineering, University of ConnecticutAbstract Machine learning has advanced the rapid prediction of inorganic materials properties, yet data scarcity for specific properties and capturing thermodynamic stability remains challenging. We propose a framework utilizing a Graph Neural Network with composition-based and crystal structure-based architectures, combined with a transfer learning scheme. This approach accurately predicts energy-related properties (e.g., total energy, energy above the convex hull, energy band gap) and data-scarce mechanical properties (e.g., bulk and shear modulus). Our model incorporates four-body interactions, capturing periodicity and structural characteristics. It outperforms state-of-the-art models in 8 materials property regression tasks. Also, this model predicts local atomic environments and global structural features better than several models. Transfer learning addresses mechanical property data scarcity, while separate architecture analysis allows application to materials lacking crystal structure information. Our framework’s interpretability aids in understanding elemental contributions, enhancing material design and discovery. Continuous advancements promise further performance improvements, driving efficient and accurate materials property prediction.https://doi.org/10.1038/s41524-024-01472-7 |
| spellingShingle | Mohammad Madani Valentina Lacivita Yongwoo Shin Anna Tarakanova Accelerating materials property prediction via a hybrid Transformer Graph framework that leverages four body interactions npj Computational Materials |
| title | Accelerating materials property prediction via a hybrid Transformer Graph framework that leverages four body interactions |
| title_full | Accelerating materials property prediction via a hybrid Transformer Graph framework that leverages four body interactions |
| title_fullStr | Accelerating materials property prediction via a hybrid Transformer Graph framework that leverages four body interactions |
| title_full_unstemmed | Accelerating materials property prediction via a hybrid Transformer Graph framework that leverages four body interactions |
| title_short | Accelerating materials property prediction via a hybrid Transformer Graph framework that leverages four body interactions |
| title_sort | accelerating materials property prediction via a hybrid transformer graph framework that leverages four body interactions |
| url | https://doi.org/10.1038/s41524-024-01472-7 |
| work_keys_str_mv | AT mohammadmadani acceleratingmaterialspropertypredictionviaahybridtransformergraphframeworkthatleveragesfourbodyinteractions AT valentinalacivita acceleratingmaterialspropertypredictionviaahybridtransformergraphframeworkthatleveragesfourbodyinteractions AT yongwooshin acceleratingmaterialspropertypredictionviaahybridtransformergraphframeworkthatleveragesfourbodyinteractions AT annatarakanova acceleratingmaterialspropertypredictionviaahybridtransformergraphframeworkthatleveragesfourbodyinteractions |