Predicting thermoelectric properties from crystal graphs and material descriptors - first application for functional materials
This work addresses material science researchers by improving prediction of functional thermoelectric properties, but it is incremental as it applies existing methods to a new domain.
The researchers tackled the problem of predicting thermoelectric properties by comparing Crystal Graph Convolutional Neural Networks (CGCNN), Fully Connected Neural Networks (FCNN), and XGBoost, finding that a three-layer FCNN using DFT-informed descriptors outperformed CGCNN and XGBoost in predicting scattering-time independent thermoelectric powerfactor.
We introduce the use of Crystal Graph Convolutional Neural Networks (CGCNN), Fully Connected Neural Networks (FCNN) and XGBoost to predict thermoelectric properties. The dataset for the CGCNN is independent of Density Functional Theory (DFT) and only relies on the crystal and atomic information, while that for the FCNN is based on a rich attribute list mined from Materialsproject.org. The results show that the optimized FCNN is three layer deep and is able to predict the scattering-time independent thermoelectric powerfactor much better than the CGCNN (or XGBoost), suggesting that bonding and density of states descriptors informed from materials science knowledge obtained partially from DFT are vital to predict functional properties.