The Graph Neural Network Model

Datos Bibliográficos

ID23345487
AutoresFranco Scarselli (0000-0003-1307-0772, University of Siena), M Gori (University of Siena), Ah Chung Tsoi Ah Chung Tsoi, Ah Chung Tsoi (0000-0003-2904-7008, Hong Kong Baptist University), Markus Hagenbuchner (0000-0002-4884-753X, University of Wollongong), Gabriele Monfardini (University of Siena)
Año2009
Volumen20
Número1
Páginas61-80
Fecha de publicación2009-01-01
Peer ReviewedSí
Open AccessSí
TipoARTICLE
RevistaIEEE Transactions on Neural Networks (JOURNAL)
Identificadores de la revistaISSN: 1045-9227 • E-ISSN: 1941-0093
EditorialInstitute of Electrical and Electronics Engineers (IEEE) (PUBLISHER)
DOI10.1109/tnn.2008.2005605
PMID19068426
OpenAlexW2116341502
IdiomaEN
Citas recibidas72
Referencias citadas60

Many underlying relationships among data in several areas of science and engineering, e.g., computer vision, molecular chemistry, molecular biology, pattern recognition, and data mining, can be represented in terms of graphs. In this paper, we propose a new neural network model, called graph neural network (GNN) model, that extends existing neural network methods for processing the data represented in graph domains. This GNN model, which can directly process most of the practically useful types of graphs, e.g., acyclic, cyclic, directed, and undirected, implements a function tau(G,n) is an element of IR(m) that maps a graph G and one of its nodes n into an m-dimensional Euclidean space. A supervised learning algorithm is derived to estimate the parameters of the proposed GNN model. The computational cost of the proposed algorithm is also considered. Some experimental results are shown to validate the proposed learning algorithm, and to demonstrate its generalization capabilities.

Artificial neural network · Graph · Advanced Graph Neural Networks · Artificial Intelligence · Computer Science · Graph Theory and Algorithms · Neural Networks and Applications · Theoretical Computer Science