Advertisement

Machine Learning

, Volume 20, Issue 1–2, pp 95–117 | Cite as

The Appropriateness of Predicate Invention as Bias Shift Operation in ILP

  • Irene Stahl
Article

Abstract

The task of predicate invention in Inductive Logic Programming is to extend the hypothesis language with new predicates if the vocabulary given initially is insufficient for the learning task. However, whether predicate invention really helps to make learning succeed in the extended language depends on the language bias currently employed.

In this paper, we investigate for which commonly employed language biases predicate invention is an appropriate shift operation. We prove that for some restricted languages predicate invention does not help when the learning task fails and we characterize the languages for which predicate invention is useful. We investigate the decidability of the bias shift problem for these languages and discuss the capabilities of predicate invention as a bias shift operation.

Inductive Logic Programming Bias Shift Predicate Invention 

References

  1. Cohen, W.. (1993). PAC-learning a restricted class of recursive logic programs. In Proc. of the 3rd International Workshop on Inductive Logic Programming.Google Scholar
  2. De Raedt, L. (1992). Interactive theory revision: an inductive logic programming approach. Academic Press.Google Scholar
  3. De Raedt, L. & Bruynooghe, M. (1989). Towards friendly concept learners. In Proc. of IJCAI.Google Scholar
  4. De Raedt, L. & Bruynooghe, M. (1992). Interactive concept-learning and constructive induction by analogy. Machine Learning, 8 (2):107–150.Google Scholar
  5. De Raedt, L., Feyaerts, J. & Bruynooghe, M. (1991). Acquiring object-knowledge for learning systems. In Y. Kodratoff, editor, Proc. of the Fifth European Working Session on Learning. Springer.Google Scholar
  6. Dzeroski, S., Muggleton, S. & Russel, S. (1992). PAC-learnability of determinate logic programs. In Proc. of the 5th ACM Workshop on Computational Learning Theory.Google Scholar
  7. Flach, P.A. (1993). Predicate invention in inductive data engineering. In Machine Learning: ECML-93, European Conference on Machine Learning, Wien, Austria. Springer.Google Scholar
  8. Gold, E.M. (1967). Language identification in the limit. Information and Control, 10:447–474.Google Scholar
  9. Kietz, J.U. & Wrobel, S. (1992). Controlling the complexity of learning in logic through syntactic and task-oriented models. In S. Muggleton, editor, Inductive Logic Programming. Academic Press.Google Scholar
  10. Kietz, J.U. & Morik, K. (1994). A polynomial approach to the constructive induction of structural knowledge. Machine Learning, 14:193–217.Google Scholar
  11. Kleene, S.C. (1952). Finite axiomatizability of theories in the predicate calculus using additional predicate symbols. In Two Papers on the Predicate Calculus, number 10 in Memoirs of the American Mathematical Society.Google Scholar
  12. Lapointe, S., Ling, C. & Matwin, S. (1993). Constructive inductive logic programming. In Proc. of the IJCAI-93. Morgan Kaufmann.Google Scholar
  13. Mitchell, T.M. (1980). The need for biases in learning generalizations. In J. W. Shavlik and T. G. Dietterich, editors, Readings in Machine Learning. Morgan Kaufmann.Google Scholar
  14. Muggleton, S. (1988). A strategy for constructing new predicates in first order logic. In Proceedings of the Third European Working Session on Learning. Pitman.Google Scholar
  15. Muggleton, S. (1990). Inductive logic programming. In First Conference on Algorithmic Learning Theory, Tokio, Ohmsha.Google Scholar
  16. Muggleton, S. (1993). Inductive logic programming: Derivations, successes and shortcoming. In Machine Learning: ECML-93, European Conference on Machine Learning, Wien, Austria. Springer.Google Scholar
  17. Muggleton, S. & Buntine, W. (1988). Machine invention of first-order predicates by inverting resolution. In Fifth International Conference on Machine Learning. Morgan Kaufmann.Google Scholar
  18. Muggleton, S. & De Raedt, L. (1994). Inductive logic programming: Theory and methods. Journal of Logic Programming, Special Issue on 10 Years of Logic Programming.Google Scholar
  19. Muggleton, S. & Feng, C. (1990). Efficient induction of logic programs. In First Conference on Algorithmic Learning Theory, Tokyo, Ohmsha.Google Scholar
  20. Quinlan, J.R. (1990). Learning logical definitions from relations. Machine Learning, 5:239–266.Google Scholar
  21. Reinhardt, K. (1993). Personal communication.Google Scholar
  22. Rouveirol, C. (1991). ITOU: Induction de Théories en Ordre Un. PhD thesis, Université Paris Sud, Centre d'Orsay.Google Scholar
  23. Rouveirol, C. (1992). ITOU: Induction of first order theories. In Muggleton, S., editor, Inductive Logic Programming. Academic Press.Google Scholar
  24. Srinivasan, A., Muggleton, S., & Bain, M. (1992). Distinguishing exceptions from noise in non-monotonic learning. In Proceedings of ILP'92, Tokyo.Google Scholar
  25. Stahl, I., Tausend, B., & Wirth, R. (1993). Two methods for improving inductive logic programming systems. In Machine Learning: ECML-93, European Conference on Machine Learning, Wien, Austria. Springer.Google Scholar
  26. Tausend, B. (1992). Using and adapting schemes for the induction of horn clauses. In ECAI Workshop Logical Approaches to Machine Learning, Wien.Google Scholar
  27. Wirth. R. (1989). Lernverfahren zur Vervollständigung von Hornklauselmengen durch inverse Resolution. Dissertation, Fakultät Informatik, Universität Stuttgart.Google Scholar
  28. Wirth, R. & O'Rorke, P. (1991). Constraints on predicate invention. In Eighth International Conference on Machine Learning. Morgan Kaufmann.Google Scholar
  29. Wrobel, S. (1994). Concept formation during interactive theory revision. Machine Learning, 14:169–191.Google Scholar
  30. Yardeni, E. & Shapiro, E. (1991). A type system for logic programs. Journal of Logic Programming, (10):125–153.Google Scholar

Copyright information

© Kluwer Academic Publishers 1995

Authors and Affiliations

  • Irene Stahl
    • 1
  1. 1.Fakultät InformalikUniversität StuttgartStuttgartGermany

Personalised recommendations