Integrating multiple materials science projects in a single neural network
Traditionally, machine learning for materials science is based on database-specific models and is limited in the number of predictable parameters. Here, a versatile graph-based neural network can integrate multiple data sources, allowing the prediction of more than 40 parameters simultaneously.
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Publishing Group
2020-07-01
|
| Series: | Communications Materials |
| Online Access: | https://doi.org/10.1038/s43246-020-00052-8 |
| id |
doaj-df9217b218ca4e14b7b2a461db83f7d6 |
|---|---|
| record_format |
Article |
| spelling |
doaj-df9217b218ca4e14b7b2a461db83f7d62021-08-01T11:15:57ZengNature Publishing GroupCommunications Materials2662-44432020-07-011111010.1038/s43246-020-00052-8Integrating multiple materials science projects in a single neural networkKan Hatakeyama-Sato0Kenichi Oyaizu1Department of Applied Chemistry, Waseda UniversityDepartment of Applied Chemistry, Waseda UniversityTraditionally, machine learning for materials science is based on database-specific models and is limited in the number of predictable parameters. Here, a versatile graph-based neural network can integrate multiple data sources, allowing the prediction of more than 40 parameters simultaneously.https://doi.org/10.1038/s43246-020-00052-8 |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Kan Hatakeyama-Sato Kenichi Oyaizu |
| spellingShingle |
Kan Hatakeyama-Sato Kenichi Oyaizu Integrating multiple materials science projects in a single neural network Communications Materials |
| author_facet |
Kan Hatakeyama-Sato Kenichi Oyaizu |
| author_sort |
Kan Hatakeyama-Sato |
| title |
Integrating multiple materials science projects in a single neural network |
| title_short |
Integrating multiple materials science projects in a single neural network |
| title_full |
Integrating multiple materials science projects in a single neural network |
| title_fullStr |
Integrating multiple materials science projects in a single neural network |
| title_full_unstemmed |
Integrating multiple materials science projects in a single neural network |
| title_sort |
integrating multiple materials science projects in a single neural network |
| publisher |
Nature Publishing Group |
| series |
Communications Materials |
| issn |
2662-4443 |
| publishDate |
2020-07-01 |
| description |
Traditionally, machine learning for materials science is based on database-specific models and is limited in the number of predictable parameters. Here, a versatile graph-based neural network can integrate multiple data sources, allowing the prediction of more than 40 parameters simultaneously. |
| url |
https://doi.org/10.1038/s43246-020-00052-8 |
| work_keys_str_mv |
AT kanhatakeyamasato integratingmultiplematerialsscienceprojectsinasingleneuralnetwork AT kenichioyaizu integratingmultiplematerialsscienceprojectsinasingleneuralnetwork |
| _version_ |
1721246155430428672 |