1)	Node-focused tasks: these tasks predict value or class for each single node or edge in the testing set. Examples include node classification, link prediction, node recommendation.

2)	Graph-focused tasks: these tasks predict value or class for the entire graph, and testing set contains multiple graphs. Examples include graph classification and graph generation. 


University of Notre Dame

In deep learning, a graph is represented as $G=(V,E)$, where $V=\{v_1,v_2,..,v_N\}$  is the set of nodes, and 	$E\subseteq V\times V$ is the set of edges between nodes. $N=|V|$ and $M=|E|$. $\mathbf{A}$ is the adjacency matrix of this graph.

Graphs Representation In Deep Learning

Zhang, Z., Cui, P., & Zhu, W. (2020). Deep learning on graphs: A survey. IEEE Transactions on Knowledge and Data Engineering.

Deep Learning on Graphs: A Survey

According to model structure and training strategies, deep learning algorithms on graphs can be divided into 5 categories:

1)	Graph Recurrent Neural Networks

2)	Graph Convolutional Networks

3)	Graph Autoencoders

4)	Graph Reinforcement Learning

5)	Graph Adversarial Methods


Learn Before

Related