Materials Property Prediction Based on Graph Neural Network and Multi-Task Learning
- 1.3 Statistics, Machine Learning, Data Assimilation, Algorithms, Mathematical Foundation, Data Mining
- 2.1 Material (Material Informatics, Remote Experiment, Laboratory Automation, etc.)
Masatoshi Hanai
Information Technology Center
Project Research Associate
Materials property prediction via computational physical simulation and machine learning is an essential task for materials discovery. This study focuses on the development of new machine learning methods based on the graph neural network and multi-task learning to effectively utilize small data sets as well as large data sets.
Research collaborators
School of Engineering, The University of Tokyo
SDGs
Contact
- Masatoshi Hanai
- Email: hanai[at]ds.itc.u-tokyo.ac.jp
※[at]=@
