Advertisement

Analogical logic program synthesis algorithm that can refute inappropriate similarities

  • Ken Sadohara
  • Makoto Haraguchi
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 997)

Abstract

This paper presents an algorithmic learning theory for analogical synthesis of logic programs from their examples. An analogical synthesizer is defined as a kind of inductive inference machine that uses analogy. More precisely speaking, it synthesizes target programs from their examples, given a source program to which the target programs should be similar. One of the difficulties in realizing an efficient analogical synthesizer is to distinguish useless and inappropriate similarities from the other. A similarity is inappropriate if every similar program with respect to the similarity is not correct. If our synthesizer cannot refute such similarities then it would waste computational resources without succeeding to find a desired program.

To cope with this hard problem on analogical synthesis, this paper first applies the notion of refutably inferable class of linear programs, and obtains a basic synthesizer. It has a function of refuting inappropriate similarities. Secondly this paper investigates another method of refuting inappropriate similarities, using an analogous technique that has been employed for theorem proving with abstraction. Incorporating this method into the basic synthesizer, we obtain a more efficient one. All the synthesizers presented in this paper are proved to identify a similar correct program in the limit, given a source program.

Keywords

Logic Program Predicate Symbol Inductive Logic Programming Source Program Correct Program 
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.
    Bipin Indurkhya. On the role of interpretive analogy in learning. In Proc. 1st Internat. Workshop on Algorithmic Learning Theory, pages 174–189, 1990.Google Scholar
  2. 2.
    Hiroki Ishizaka. Model inference incorporating generalization. Journal of Information Processing, 11(3):206–211, 1988.Google Scholar
  3. 3.
    J.W. Lloyd. Foundations of Logic Programming. Springer-Verlag, second edition, 1987.Google Scholar
  4. 4.
    Stephen Muggleton. Inductive logic programming. In Inductive Logic Programming, pages 3–27. ACADEMIC PRESS, 1992.Google Scholar
  5. 5.
    Yasuhito Mukouchi and Setsuo Arikawa. Towards a mathematical theory of machine discovery from facts. Theoretical Computer Science, 137:53–84, 1995.Google Scholar
  6. 6.
    David A. Plaisted. Theorem proving with abstraction. Artificial Intelligence, 16:47–108, 1981.CrossRefGoogle Scholar
  7. 7.
    G.D. Plotkin. A note on inductive generalization. In Machine Intelligence 5, pages 153–163. Edinburgh University Press, 1970.Google Scholar
  8. 8.
    Luc De Raedt and Maurice Bruynooghe. Interactive concept-learning and constructive induction by analogy. Machine Learning, 8:107–150, 1992.Google Scholar
  9. 9.
    J.C. Reynolds. Transformational systems and the algebraic structure of atomic formulas. In Machine Intelligence 5, pages 135–153. Edinburgh University Press, 1970.Google Scholar
  10. 10.
    J. A. Robinson. A machine-oriented logic based on the resolution principle. Journal of the Association for Computing Machinery, 12:13–41, Mach. 1965.Google Scholar
  11. 11.
    Céline Rouveirol. Extension of inversion of resolution applied to theory completion. In Inductive Logic Programming, pages 64–92. ACADEMIC PRESS, 1992.Google Scholar
  12. 12.
    Ken Sadohara and Makoto Haraguchi. Analogical logic program synthesis from examples. In Proc. 8th European Conference on Machine Learning, Lecture Notes in Artificial Intelligence Vol. 914, pages 232–244. Springer-Verlag, 1995.Google Scholar
  13. 13.
    Seiichiro Sakurai and Makoto Haraguchi. Towards learning by abstraction. In Proc. 2nd Internat. Workshop on Algorithmic Learning Theory, pages 288–298, 1991.Google Scholar
  14. 14.
    Ehud Y. Shapiro. Inductive inference of theories from facts. Technical Report 192, Yale University Computer Science Dept., 1981.Google Scholar
  15. 15.
    Birgit Tausend and Siegfried Bell. Analogical reasoning for logic programming. In Inductive Logic Programming, pages 397–408. ACADEMIC PRESS, 1992.Google Scholar
  16. 16.
    Patrick R.J. van der Laag and Shan-Hwei Nienhuys-Cheng. Subsumption and refinement in model inference. In Proc. 6th European Conference on Machine Learning, pages 95–114. Springer-Verlag, 1993.Google Scholar
  17. 17.
    Akihiro Yamamoto. Procedural semantics and negative information of elementary formal system. J. Logic Programming, 13:89–97, 1992.MathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Ken Sadohara
    • 1
  • Makoto Haraguchi
    • 2
  1. 1.Department of Systems Science Tokyo Institute of TechnologyMidori-ku YokohamaJapan
  2. 2.Division of Electronics and Information EngineeringHokkaido UniversitySapporoJapan

Personalised recommendations