Predicting Material Properties Using a 3D Graph Neural Network with Invariant Local Descriptors
This work addresses the need for accurate material property prediction in materials science, particularly for applications like gas adsorption and ion conductivity, though it appears incremental as it builds on existing graph neural network methods.
The paper tackled the problem of predicting material properties by developing a 3D graph neural network that models atomic interactions in three-dimensional space, outperforming existing graph-based models on challenging datasets like Henry's constant for gas adsorption in MOFs and ion conductivity in solid-state crystals.
Accurate prediction of physical properties is critical for discovering and designing novel materials. Machine learning technologies have attracted significant attention in the materials science community for their potential for large-scale screening. Graph Convolution Neural Network (GCNN) is one of the most successful machine learning methods because of its flexibility and effectiveness in describing 3D structural data. Most existing GCNN models focus on the topological structure but overly simplify the three-dimensional geometric structure. However, in materials science, the 3D-spatial distribution of atoms is crucial for determining the atomic states and interatomic forces. This paper proposes an adaptive GCNN with a novel convolution mechanism that simultaneously models atomic interactions among all neighbor atoms in three-dimensional space. We apply the proposed model to two distinctly challenging problems on predicting material properties. The first is Henry's constant for gas adsorption in Metal-Organic Frameworks (MOFs), which is notoriously difficult because of its high sensitivity to atomic configurations. The second is the ion conductivity in solid-state crystal materials, which is difficult because of few labeled data available for training. The new model outperforms existing graph-based models on both data sets, suggesting that the critical three-dimensional geometric information is indeed captured.