MC-TopLog: Complete Multi-clause Learning Guided by a Top Theory

  • Stephen H. Muggleton
  • Dianhuan Lin
  • Alireza Tamaddoni-Nezhad
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7207)

Abstract

Within ILP much effort has been put into designing methods that are complete for hypothesis finding. However, it is not clear whether completeness is important in real-world applications. This paper uses a simplified version of grammar learning to show how a complete method can improve on the learning results of an incomplete method. Seeing the necessity of having a complete method for real-world applications, we introduce a method called ⊤-directed theory co-derivation, which is shown to be correct (ie. sound and complete). The proposed method has been implemented in the ILP system MC-TopLog and tested on grammar learning and the learning of game strategies. Compared to Progol5, an efficient but incomplete ILP system, MC-TopLog has higher predictive accuracies, especially when the background knowledge is severely incomplete.

Keywords

Background Knowledge Logic Program Hypothesis Space Common Generalisation Recursive Theory 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Blockeel, H., De Raedt, L.: Top-down induction of first order logical decision trees. Artificial Intelligence 101(1-2), 285–297 (1998)MathSciNetMATHCrossRefGoogle Scholar
  2. 2.
    Boström, H., Idestam-Almquist, P.: Induction of logic programs by example-guided unfolding. The Journal of Logic Programming 40, 159–183 (1999)MATHCrossRefGoogle Scholar
  3. 3.
    Bratko, I.: Refining Complete Hypotheses in ILP. In: Džeroski, S., Flach, P.A. (eds.) ILP 1999. LNCS (LNAI), vol. 1634, pp. 44–55. Springer, Heidelberg (1999)CrossRefGoogle Scholar
  4. 4.
    Cohen, W.: Grammatically biased learning: Learning logic programs using an explicit antecedent description language. Artificial Intelligence 68, 303–366 (1994)MATHCrossRefGoogle Scholar
  5. 5.
    Inoue, K.: Induction as consequence finding. Machine Learning 55, 109–135 (2004)MATHCrossRefGoogle Scholar
  6. 6.
    Kedar-Cabelli, S.T., McCarty, L.T.: Explanation-based generalization as resolution theorem proving. In: Proceedings of ICML 1987, pp. 383–389. Morgan Kaufmann, Los Altos (1987)Google Scholar
  7. 7.
    Kimber, T., Broda, K., Russo, A.: Induction on Failure: Learning Connected Horn Theories. In: Erdem, E., Lin, F., Schaub, T. (eds.) LPNMR 2009. LNCS, vol. 5753, pp. 169–181. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  8. 8.
    Lin, D.: Efficient, complete and declarative search in inductive logic programming. Master’s thesis, Imperial College London (September 2009)Google Scholar
  9. 9.
    Lin, D., Chen, J., Watanabe, H., Muggleton, S.H., Jain, P., Sternberg, M., Baxter, C., Currie, R., Dunbar, S., Earll, M., Salazar, D.: Does Multi-clause Learning Help in Real-world Applications? In: Muggleton, S.H., Tamaddoni-Nezhad, A., Lisi, F.A. (eds.) ILP 2011. LNCS (LNAI), vol. 7207, pp. 222–238. Springer, Heidelberg (2012)Google Scholar
  10. 10.
    Malerba, D.: Learning recursive theories in the normal ILP setting. Fundamenta Informaticae 57, 39–77 (2003)MathSciNetMATHGoogle Scholar
  11. 11.
    Muggleton, S.H.: Inverse entailment and Progol. New Generation Computing 13, 245–286 (1995)CrossRefGoogle Scholar
  12. 12.
    Muggleton, S.H., Bryant, C.H.: Theory Completion Using Inverse Entailment. In: Cussens, J., Frisch, A.M. (eds.) ILP 2000. LNCS (LNAI), vol. 1866, pp. 130–146. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  13. 13.
    Muggleton, S.H., Feng, C.: Efficient induction of logic programs. In: ALT 1990, pp. 368–381. Ohmsha, Tokyo (1990)Google Scholar
  14. 14.
    Muggleton, S.H., Santos, J.C.A., Tamaddoni-Nezhad, A.: TopLog: ILP Using a Logic Program Declarative Bias. In: Garcia de la Banda, M., Pontelli, E. (eds.) ICLP 2008. LNCS, vol. 5366, pp. 687–692. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  15. 15.
    Muggleton, S., Santos, J., Tamaddoni-Nezhad, A.: ProGolem: A System Based on Relative Minimal Generalisation. In: De Raedt, L. (ed.) ILP 2009. LNCS, vol. 5989, pp. 131–148. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  16. 16.
    Muggleton, S.H., Xu, C.: Can ILP learn complete and correct game strategies? In: Late-breaking Proceedings of ILP. Imperial College London Press (2011)Google Scholar
  17. 17.
    Plotkin, G.D.: Automatic Methods of Inductive Inference. PhD thesis, Edinburgh University (August 1971)Google Scholar
  18. 18.
    De Raedt, L., Dehaspe, L.: Clausal discovery. Machine Learning 26, 99–146 (1997)MATHCrossRefGoogle Scholar
  19. 19.
    De Raedt, L., Lavrac, N., Dzeroski, S.: Multiple predicate learning. In: IJCAI, pp. 1037–1043 (1993)Google Scholar
  20. 20.
    Ray, O.: Nonmonotonic abductive inductive learning. Journal of Applied Logic 7(3), 329–340 (2009)MathSciNetMATHCrossRefGoogle Scholar
  21. 21.
    Reynolds, J.C.: Transformational systems and the algebraic structure of atomic formulas. In: Meltzer, B., Michie, D. (eds.) Machine Intelligence, vol. 5, pp. 135–151. Edinburgh University Press, Edinburgh (1969)Google Scholar
  22. 22.
    Yamamoto, A.: Which Hypotheses can be Found with Inverse Entailment? In: Džeroski, S., Lavrač, N. (eds.) ILP 1997. LNCS, vol. 1297, pp. 296–308. Springer, Heidelberg (1997)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Stephen H. Muggleton
    • 1
  • Dianhuan Lin
    • 1
  • Alireza Tamaddoni-Nezhad
    • 1
  1. 1.Department of ComputingImperial College LondonUK

Personalised recommendations