Feature engineering descriptors, transforms, and machine learning for grain boundaries and variable-sized atom clusters
Abstract Obtaining microscopic structure-property relationships for grain boundaries is challenging due to their complex atomic structures. Recent efforts use machine learning to derive these relationships, but the way the atomic grain boundary structure is represented can have a significant impact...
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/s41524-024-01509-x |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832585595427225600 |
|---|---|
| author | C. Braxton Owens Nithin Mathew Tyce W. Olaveson Jacob P. Tavenner Edward M. Kober Garritt J. Tucker Gus L. W. Hart Eric R. Homer |
| author_facet | C. Braxton Owens Nithin Mathew Tyce W. Olaveson Jacob P. Tavenner Edward M. Kober Garritt J. Tucker Gus L. W. Hart Eric R. Homer |
| author_sort | C. Braxton Owens |
| collection | DOAJ |
| description | Abstract Obtaining microscopic structure-property relationships for grain boundaries is challenging due to their complex atomic structures. Recent efforts use machine learning to derive these relationships, but the way the atomic grain boundary structure is represented can have a significant impact on the predictions. Key steps for property prediction common to grain boundaries and other variable-sized atom clustered structures include: (1) describing the atomic structure as a feature matrix, (2) transforming the variable-sized feature matrix to a fixed length common to all structures, and (3) applying a machine learning algorithm to predict properties from the transformed matrices. We examine how these steps and different combinations of engineered features impact the accuracy of grain boundary energy predictions using a database of over 7000 grain boundaries. Additionally, we assess how different engineered features support interpretability, offering insights into the physics of the structure-property relationships. |
| format | Article |
| id | doaj-art-cabde7a42428430bacac08c9f4cac5ef |
| 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-cabde7a42428430bacac08c9f4cac5ef2025-01-26T12:43:03ZengNature Portfolionpj Computational Materials2057-39602025-01-0111111510.1038/s41524-024-01509-xFeature engineering descriptors, transforms, and machine learning for grain boundaries and variable-sized atom clustersC. Braxton Owens0Nithin Mathew1Tyce W. Olaveson2Jacob P. Tavenner3Edward M. Kober4Garritt J. Tucker5Gus L. W. Hart6Eric R. Homer7Department of Computer Science, Brigham Young UniversityGroup T-1, Theoretical Division, Los Alamos National LaboratoryDepartment of Physics and Astronomy, Brigham Young UniversityKBR, Inc., Intelligent Systems Division, NASA Ames Research CenterGroup T-1, Theoretical Division, Los Alamos National LaboratoryDepartment of Physics, Baylor UniversityDepartment of Physics and Astronomy, Brigham Young UniversityDepartment of Mechanical Engineering, Brigham Young UniversityAbstract Obtaining microscopic structure-property relationships for grain boundaries is challenging due to their complex atomic structures. Recent efforts use machine learning to derive these relationships, but the way the atomic grain boundary structure is represented can have a significant impact on the predictions. Key steps for property prediction common to grain boundaries and other variable-sized atom clustered structures include: (1) describing the atomic structure as a feature matrix, (2) transforming the variable-sized feature matrix to a fixed length common to all structures, and (3) applying a machine learning algorithm to predict properties from the transformed matrices. We examine how these steps and different combinations of engineered features impact the accuracy of grain boundary energy predictions using a database of over 7000 grain boundaries. Additionally, we assess how different engineered features support interpretability, offering insights into the physics of the structure-property relationships.https://doi.org/10.1038/s41524-024-01509-x |
| spellingShingle | C. Braxton Owens Nithin Mathew Tyce W. Olaveson Jacob P. Tavenner Edward M. Kober Garritt J. Tucker Gus L. W. Hart Eric R. Homer Feature engineering descriptors, transforms, and machine learning for grain boundaries and variable-sized atom clusters npj Computational Materials |
| title | Feature engineering descriptors, transforms, and machine learning for grain boundaries and variable-sized atom clusters |
| title_full | Feature engineering descriptors, transforms, and machine learning for grain boundaries and variable-sized atom clusters |
| title_fullStr | Feature engineering descriptors, transforms, and machine learning for grain boundaries and variable-sized atom clusters |
| title_full_unstemmed | Feature engineering descriptors, transforms, and machine learning for grain boundaries and variable-sized atom clusters |
| title_short | Feature engineering descriptors, transforms, and machine learning for grain boundaries and variable-sized atom clusters |
| title_sort | feature engineering descriptors transforms and machine learning for grain boundaries and variable sized atom clusters |
| url | https://doi.org/10.1038/s41524-024-01509-x |
| work_keys_str_mv | AT cbraxtonowens featureengineeringdescriptorstransformsandmachinelearningforgrainboundariesandvariablesizedatomclusters AT nithinmathew featureengineeringdescriptorstransformsandmachinelearningforgrainboundariesandvariablesizedatomclusters AT tycewolaveson featureengineeringdescriptorstransformsandmachinelearningforgrainboundariesandvariablesizedatomclusters AT jacobptavenner featureengineeringdescriptorstransformsandmachinelearningforgrainboundariesandvariablesizedatomclusters AT edwardmkober featureengineeringdescriptorstransformsandmachinelearningforgrainboundariesandvariablesizedatomclusters AT garrittjtucker featureengineeringdescriptorstransformsandmachinelearningforgrainboundariesandvariablesizedatomclusters AT guslwhart featureengineeringdescriptorstransformsandmachinelearningforgrainboundariesandvariablesizedatomclusters AT ericrhomer featureengineeringdescriptorstransformsandmachinelearningforgrainboundariesandvariablesizedatomclusters |