![]() ![]() Through multiple layers of graph convolution updating node features based on their local chemical environment, these models can implicitly represent many-body interactions. A common theme is the use of elemental properties as node features and interatomic distances and/or bond valences as edge features. ![]() ![]() This family of models represents a molecule or crystalline material as a graph with one node for each constituent atom and edges corresponding to interatomic bonds. There has been rapid progress in the development of GNN architectures for predicting material properties such as SchNet 10, Crystal Graph Convolutional Neural Networks (CGCNN) 15, MatErials Graph Network (MEGNet) 16, improved Crystal Graph Convolutional Neural Networks (iCGCNN) 17, OrbNet 18, and similar variants 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31. From a quantum chemistry point of view, GNNs provide a unique opportunity to predict properties of solids, molecules, and proteins in a much faster way rather than by solving the computationally expensive Schrodinger equation 11, 12, 13, 14. ![]() Common applications of GNNs include community detection and link prediction in social networks 5, 6, functional time series on brain structures 7, gene DNA on regulatory networks 8, information flow through telecommunications networks 9, and property prediction for molecular and solid materials 10. Graph neural networks (GNN) 3, 4 have immense potential for modeling complex phenomena. Graphs are a powerful non-Euclidean data structure method for establishing relationships between features (nodes) and their relationships (edges) 1, 2. ALIGNN can outperform some previously reported GNN models on atomistic prediction tasks with better or comparable model training speed. We ALIGNN models for predicting 52 solid-state and molecular properties available in the JARVIS-DFT, Materials project, and QM9 databases. We demonstrate that angle information can be explicitly and efficiently included, leading to improved performance on multiple atomistic prediction tasks. We present an Atomistic Line Graph Neural Network (ALIGNN), a GNN architecture that performs message passing on both the interatomic bond graph and its line graph corresponding to bond angles. Furthermore, many material properties are known to be sensitive to slight changes in bond angles. While most existing GNN models for atomistic predictions are based on atomic distance information, they do not explicitly incorporate bond angles, which are critical for distinguishing many atomic structures. Graph neural networks (GNN) have been shown to provide substantial performance improvements for atomistic material representation and modeling compared with descriptor-based machine learning models. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. Archives
December 2022
Categories |