Predicate invention from a few examples

  • Riverson Rios
  • Stan Matwin
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1418)


One main difficulty of Inductive Logic Programming lies in learning recursively defined predicates. Today's systems strongly rely on a set of supporting predicates known as the background knowledge, which assumes that the user knows in advance what sort of predicates are required by the target definition. Predicate invention can remedy the situation by extending the specification language with new concepts that appear neither in the examples nor in the background knowledge, and finding a definition for them. A serious concern is that no examples of the invented predicate are explicitly given but rather of the target predicate, so learning has to be done in the absence or scarcity of examples. This work shows an autonomous learning method based on inverting clausal implication that can invent the recursive predicates it needs. The learner is endowed with means to work from a small data set.


learning inductive logic programming 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Aha, D.W., Lapointe, S., Ling, C. & Matwin, S. (1994). Inverting implication with small training sets. In Proc. ECML (31–48). Catania, ItalyGoogle Scholar
  2. [2]
    Feng, C. (1992). Inducing temporal fault diagnostic rules from a qualitative model. In Inductive Logic Programming (473–493). San Diego, CAGoogle Scholar
  3. [3]
    Bergadano, F. & Gunetti, D. (1993). An interactive system to learn functional logic programs. In Proc. 13th IJCAI (1044–49). Chambéry, FranceGoogle Scholar
  4. [4]
    Cohen, W. (1995). Pac-learning recursive logic programs: negative results. Journal of Artificial Intelligence Research 2:541–573Google Scholar
  5. [5]
    Idestam-Almquist, P. (1993). Generalization of Clauses. Ph.D. Thesis. Dept. of Computer and System Sciences. Stockholm University. Stockholm, SwedenGoogle Scholar
  6. [6]
    Idestam-Almquist, P. (1996). Efficient induction of recursive definitions by structural analysis of saturation. In (L. De Raedt, ed.) Advances in ILP (192–205). Amsterdam, The NetherlandsGoogle Scholar
  7. [7]
    Kijsirikul, B., Numao, M. & Shimura, M. (1992). Discrimination-based constructive induction of logic programs. In Proc. 10 th National Conference on Artificial Intelligence (44–49). San José, CAGoogle Scholar
  8. [8]
    Lapointe, S., Ling, C. & Matwin, S. (1993). Constructive inductive logic programming. In Proc. of the 13 th IJCAI (273–281). Chambéry, FranceGoogle Scholar
  9. [9]
    Lapointe, S. & Matwin, S. (1992). Sub-unification: A tool for efficient induction of recursive programs. In Proc. 9th ICML (273–281). Aberdeen, ScotlandGoogle Scholar
  10. [10]
    Lavrac, N & Dzeroski, S. (1994). Inductive Logic Programming: Techniques and Applications. Ellis HorwoodGoogle Scholar
  11. [11]
    Ling, C. (1991). Inductive learning from good examples. In (J. Mylopoulos, R. Reiter, eds.) Proc. 12 th IJCAI (751–756). Sydney, AustraliaGoogle Scholar
  12. [12]
    Marcinkowski, J. & Pacholski, L. (1992). Undecidability of the Horn clause implication problem. In Proc. 33rd IEEE Symposium on the Foundations of Computer Science (354-362). Pittsburgh, PAGoogle Scholar
  13. [13]
    Martin. L. & Vrain, C. (1997). Systematic predicate invention in inductive logic programming. In Proc. Seventh International Workshop on ILP. Prague, Czech RepublicGoogle Scholar
  14. [14]
    Mofizur, C. & Numao, C. (1996). Top-down induction of recursive programs from small number of sparse examples. In (L. De Raedt, ed.) Advances in ILP (236–253). Amsterdam, The NetherlandsGoogle Scholar
  15. [15]
    Muggleton, S. (1994). Inductive logic programming: derivations, successes and shortcomings. SIGART Bulletin 5(1):5–11Google Scholar
  16. [16]
    Muggleton, S., King, R. & Sternberg, M. (1992). Protein secondary structure prediction using logic. In Proc. 2nd Int'l Workshop on ILP. Tokyo, JapanGoogle Scholar
  17. [17]
    Plotkin, G.D. (1971). Automatic Methods of Inductive Inference. Ph.D. Thesis. Edinburgh University, Edinburgh, ScotlandGoogle Scholar
  18. [18]
    Rios, R. & Matwin, S. (1996). Efficient induction of recursive Prolog definitions. In Proc. 11 th Canadian Artificial Intelligence Conference (240–248). Toronto, CanadaGoogle Scholar
  19. [19]
    Rios, R. (1998). Learning Recursive Definitions in Prolog. Ph.D. Thesis. School of Information Technology and Engineering, University of Ottawa. Ottawa, CanadaGoogle Scholar
  20. [20]
    Schmidt-Schaug, M. (1988). Implication of clauses is undecidable. Theoretical Computer Science 59:287–296CrossRefGoogle Scholar
  21. [21]
    Stahl, I. (1996). Predicate invention in inductive logic programming. In (L. De Raedt, ed.) Advances in ILP (34–47). Amsterdam, The NetherlandsGoogle Scholar
  22. [22]
    Stahl, I., Weber, I. (1994). The arguments of newly invented predicates in ILP. In Proc. 4th Int'l Workshop on ILP. Bad Honnef/Bonn, GermanyGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Riverson Rios
    • 1
  • Stan Matwin
    • 1
  1. 1.School of Information Technology and EngineeringUniversity of OttawaCanada

Personalised recommendations