A new model for learning in graph domains

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.

I. Introduction

In several machine learning applications the data of interest can be suitably represented in form of sequences, trees, and, more generally, directed or undirected graphs, f.i. in chemics [1], software engineering, image processing [2]. In those applications, the goal consists of learning from examples a function that maps a graph and one of its nodes to a vector of reals: .