Abstract

Stochastic Constraint Programming is an extension of Constraint Programming for modelling and solving combinatorial problems involving uncertainty. A solution to such a problem is a policy tree that specifies decision variable assignments in each scenario. Several complete solution methods have been proposed, but the authors recently showed that an incomplete approach based on neuroevolution is more scalable. In this paper we hybridise neuroevolution with constraint filtering on hard constraints, and show both theoretically and empirically that the hybrid can learn more complex policies more quickly.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Majercik, S.M.: Stochastic Boolean Satisfiability. In: Handbook of Satisfiability, ch. 27, pp. 887–925. IOS Press, Amsterdam (2009)Google Scholar
  2. 2.
    Minsky, M., Papert, S.: Perceptrons: An Introduction to Computational Geometry. The MIT Press, Cambridge (1972)Google Scholar
  3. 3.
    Prestwich, S.D., Tarim, S.A., Rossi, R., Hnich, B.: Evolving Parameterised Policies for Stochastic Constraint Programming. In: Gent, I.P. (ed.) CP 2009. LNCS, vol. 5732, pp. 684–691. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  4. 4.
    Racca, R.: Can Periodic Perceptrons Replace Multi-Layer Perceptrons? Pattern Recognition Letters 21, 1019–1025 (2000)MATHCrossRefGoogle Scholar
  5. 5.
    Régin, J.-C.: A Filtering Algorithm for Constraints of Difference in CSPs. In: 12th National Conference on Artificial Intelligence, pp. 362–367. AAAI Press, Menlo Park (1994)Google Scholar
  6. 6.
    Stanley, K.O., Miikkulainen, R.: A Taxonomy for Artificial Embryogeny. Artificial Life 9(2), 93–130 (2003)CrossRefGoogle Scholar
  7. 7.
    Walsh, T.: Stochastic Constraint Programming. In: 15th European Conference on Artificial Intelligence (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Steve D. Prestwich
    • 1
  • S. Armagan Tarim
    • 2
  • Roberto Rossi
    • 3
  • Brahim Hnich
    • 4
  1. 1.Cork Constraint Computation CentreUniversity College CorkIreland
  2. 2.Department of ManagementHacettepe UniversityAnkaraTurkey
  3. 3.Logistics, Decision and Information Sciences GroupWageningen URThe Netherlands
  4. 4.Faculty of Computer ScienceIzmir University of EconomicsTurkey

Personalised recommendations