Siamese Equivariant Neural Network for Property Predictions in Defect Materials

Defect properties such as formation energy using density functional theory (DFT) in defect materials is critical to understanding their properties and defect growth mechanisms, yet the accurate and efficient calculation of them remains a challenge in materials science. In this study, we introduce the Siamese Equivariant Neural Network (SENN) for predicting properties in defect material systems. We leverage E(3) equivariance to construct representations for both defects and their host crystal structures, and use the difference of the learned representation for predictions, thereby forming a Siamese network structure. Our results demonstrate that the E(3) model surpasses previous invariant graph neural network models, and the proposed SENN further enhances performance. Our model outperforms invariant graph neural networks and pure equivariant neural networks on various defect material datasets. Our model can be applied for fast prediction of defect properties like formation energy, which can be used for fast screening of defect material candidate systems at a large scale.