Polynomial-time MAT learning of multilinear logic programs

  • Kimihito Ito
  • Akihiro Yamamoto
Technical Papers Learning via Query
Part of the Lecture Notes in Computer Science book series (LNCS, volume 743)


In this paper we give a MAT(Minimally Adequate Teacher) learning algorithm of multilinear logic programs. MAT learning is to infer the unknown model M U that the teacher has, with membership queries and equivalence queries. In the class of multilinear programs, we show some programs which have not been proved to be MAT learnable previously. If a multilinear program P U represents M U that the teacher has, then the total running time of our learning algorithm is bounded by a polynomial in m, w and h, where m is the number of predicates in P U , h is the number of non-linear clauses in P U , and w is a parameter depending on counter-examples to equivalence queries. We also show multilinear programs with outputs are MAT learnable by extending the algorithm.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Angluin, D.: Learning regular sets from queries and counter-examples, Information and Computation, 75:87–106, 1987.CrossRefGoogle Scholar
  2. [2]
    Brainerd, W.S.: The Minimalization of Tree Automata, Information and Control, 13:483–491, 1968.Google Scholar
  3. [3]
    Ishizaka, H.: Learning Simple Deterministic Languages, in Proceedings of the Second Annual Workshop on Computional Learning Theory:162–174, 1989.Google Scholar
  4. [4]
    Ito, K.: Logic Programs which are MAT Learnable, Graduation Thesis, Faculty of Engineering, Hokkaido University, 1992.Google Scholar
  5. [5]
    Lloyd, J. W.: Foundations of Logic Programming, Springer-Verlag,1984.Google Scholar
  6. [6]
    Marque-Pucheu, G.: Rational Set of Trees and the Algebraic Semantics of Logic Programming, Acta Informatica, 20:249–260, 1983.Google Scholar
  7. [7]
    Muggleton, S.: Inductive Logic Programming, in Proceedings of the First International Workshop on Algorithmic Learning Theory:42–62, 1990.Google Scholar
  8. [8]
    Sakakibara, Y.: Learning Context-free Grammars from Structural Data in Polynomial Time, in Proceedings of 1st Workshop on Computational Learning Theory:296–310, 1988.Google Scholar
  9. [9]
    Sakakibara, Y.: Inductive Inference of Logic Programs Based on Algebraic Semantics, New Generation Computing, 7: 365–380, 1990.Google Scholar
  10. [10]
    Sakakibara, Y., Nishino, T.: EXACT Learning — Concept Learning from Queries (in Japanese), Journal of Information Processing Society Japan, 32:246–255, 1991.Google Scholar
  11. [11]
    Shapiro, E. Y.: Inductive Inference of Theories from Facts, Technical Report, 192, Yale University Computer Science Dept. 1981.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1993

Authors and Affiliations

  • Kimihito Ito
    • 1
  • Akihiro Yamamoto
    • 1
  1. 1.Department of Electrical EngineeringHokkaido UniversitySapporoJapan

Personalised recommendations