Fig. 2: Schematic representations of an atomic structure as a graph.
From: Recent advances and applications of deep learning methods in materials science
a CGCNN model in which crystals are converted to graphs with nodes representing atoms in the unit cell and edges representing atom connections. Nodes and edges are characterized by vectors corresponding to the atoms and bonds in the crystal, respectively [Reprinted with permission from ref. 67 Copyright 2019 American Physical Society], b ALIGNN65 model in which the convolution layer alternates between message passing on the bond graph and its bond-angle line graph. c MEGNet in which the initial graph is represented by the set of atomic attributes, bond attributes and global state attributes [Reprinted with permission from ref. 33 Copyright 2019 American Chemical Society] model, d iCGCNN model in which multiple edges connect a node to neighboring nodes to show the number of Voronoi neighbors [Reprinted with permission from ref. 122 Copyright 2019 American Physical Society].
