A New Neural Network Model for Contextual Processing of Graphs

  • Alessio Micheli
  • Antonio S. Sestito
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3931)


We propose a novel simple approach to deal with fairly general graph structures by neural networks. Using a constructive approach, the model Neural Network for Graphs (NN4G) exploits the contextual information stored in the hidden units progressively added to the network, without introducing cycles in the definition of the state variables. In contrast to previous neural networks for structures, NN4G is not recursive but uses standard neurons (with no feedbacks) that traverse each graph without hierarchical assumptions on its topology, allowing the extension of structured domain to cyclic directed/undirected graphs. Initial experimental results, obtained on the prediction of the boiling point of alkanes and on the classification of artificial cyclic structures, show the effectiveness of this new approach.


Singular Value Decomposition Neural Network Model Hide Unit Output Unit Maximum Absolute Error 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bianchini, M., Gori, M., Scarselli, F.: Recursive processing of cyclic graphs. In: Proc. of WCCI-IJCNN 2002, vol. 1, pp. 154–159 (2002)Google Scholar
  2. 2.
    Bianucci, A.M., Micheli, A., Sperduti, A., Starita, A.: Application of cascade correlation networks for structures to chemistry. Appl. Intell. 12, 117–146 (2000)CrossRefGoogle Scholar
  3. 3.
    Fahlman, S.E., Lebiere, C.: The cascade-correlation learning architecture. Technical Report CMU-CS-90-100, Carnegie Mellon (August 1990)Google Scholar
  4. 4.
    Frasconi, P., Gori, M., Küchler, A., Sperduti, A.: From sequences to data structures: Theory and applications. In: Kolen, J.F., Kremer, S.C. (eds.) A Field Guide to Dynamical Recurrent Networks, ch. 19, IEEE Press, Inc., Los Alamitos (2001)Google Scholar
  5. 5.
    Hammer, B., Steil, J.J.: Tutorial: Perspectives on learning with rnns. In: Proc. of ESANN 2002, pp. 357–368. D-side (2002)Google Scholar
  6. 6.
    Micheli, A., Portera, F., Sperduti, A.: A preliminary empirical comparison of recursive neural networks and tree kernel methods on regression tasks for tree structured domains. Neurocomputing 64, 73–92 (2005)CrossRefGoogle Scholar
  7. 7.
    Micheli, A., Sona, D., Sperduti, A.: Contextual processing of structured data by recursive cascade correlation. IEEE Trans. Neural Networks 15(6), 1396–1410 (2004)CrossRefGoogle Scholar
  8. 8.
    Press, W.H., Teukolsky, S.A., Vetterling, W.T., Flannery, B.P.: Numerical Recipes in C: the Art of Scientific Computing, 2nd edn. Cambridge University Press, Cambridge (1992)zbMATHGoogle Scholar
  9. 9.
    Sperduti, A., Starita, A.: Supervised neural networks for the classification of structures. IEEE Trans. Neural Networks 8(3), 714–735 (1997)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Alessio Micheli
    • 1
  • Antonio S. Sestito
    • 1
  1. 1.Dipartimento di InformaticaUniversità di PisaPisaItaly

Personalised recommendations