A new model for learning in graph domains

TLDR

Graph data are naturally represented as graphs, but traditional preprocessing into flat vectors can discard topological information and make results highly dependent on the preprocessing stage. This paper introduces a graph neural network (GNN) that processes graphs directly. The GNN extends recursive neural networks, supports directed, undirected, labelled, and cyclic graphs, and includes a learning algorithm. Experiments demonstrate the model’s properties and validate the proposed learning algorithm.

Abstract

In several applications the information is naturally represented by graphs. Traditional approaches cope with graphical data structures using a preprocessing phase which transforms the graphs into a set of flat vectors. However, in this way, important topological information may be lost and the achieved results may heavily depend on the preprocessing stage. This paper presents a new neural model, called graph neural network (GNN), capable of directly processing graphs. GNNs extends recursive neural networks and can be applied on most of the practically useful kinds of graphs, including directed, undirected, labelled and cyclic graphs. A learning algorithm for GNNs is proposed and some experiments are discussed which assess the properties of the model.