Graph neural network/Graph embedding code.

• GNN: gcn, graphSAGE
• GE: node2vec

GCN: learn a function of signals/features on a graph G = (V, E) which takes as input:

• A feature description x_i for every node i summarized in a N x D feature matrix X. (N : number of nodes, D : number of input features)
• A representative description of the graph structure in matrix form, eg: adjacency matrix A.

and produces a node-level output Z (an N x F feature matrix, where F is the number of output features per node). Graph-level outputs can be modeled by producing some form of pooling operation.

Every neural network layer can then be written as a non-linear function, consider the following simple form of a layer-wise propagation rule:

with H⁽⁰⁾=X and H^(L)=Z (z for graph-level outputs), L being the number of layers. The specific models then differ only in how is chosen and parameterized.

• Semi-Supervised Classification with Graph Convolutional Networks
• Inductive Representation Learning on Large Graphs
• node2vec: Scalable Feature Learning for Networks