Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties

TL;DR

This paper introduces a crystal graph convolutional neural network framework that accurately predicts material properties directly from crystal structures, offering interpretability and insights for materials design.

Contribution

The authors develop a universal, interpretable neural network model that directly learns from crystal structures without manual feature engineering, improving prediction accuracy across diverse materials.

Findings

Achieved high accuracy in predicting DFT-calculated properties for various crystal types.

Provided interpretability by extracting local chemical environment contributions.

Demonstrated application in discovering empirical rules for materials design.

Abstract

The use of machine learning methods for accelerating the design of crystalline materials usually requires manually constructed feature vectors or complex transformation of atom coordinates to input the crystal structure, which either constrains the model to certain crystal types or makes it difficult to provide chemical insights. Here, we develop a crystal graph convolutional neural networks framework to directly learn material properties from the connection of atoms in the crystal, providing a universal and interpretable representation of crystalline materials. Our method provides a highly accurate prediction of density functional theory calculated properties for eight different properties of crystals with various structure types and compositions after being trained with $10^4$ data points. Further, our framework is interpretable because one can extract the contributions from local chemical environments to global properties. Using an example of perovskites, we show how this information can be utilized to discover empirical rules for materials design.