Advertisement

Abduction for explanation-based learning

  • Béatrice Duval
Part 5: Inversion Of Resolution
Part of the Lecture Notes in Computer Science book series (LNCS, volume 482)

Abstract

Explanation-based Learning has raised many studies in Machine Learning community but the original process proposed by Mitchell suffers of a major drawback, the necessity to deal with a complete, correct and tractable theory, since learning results from a complete explanation, that means a complete proof, of a training instance. In this paper, we propose to learn when only a partial explanation can be found. Our idea is to extend the resolver that explains examples so that plausible completions of partial proofs can be proposed to the user. To achieve such completions, abductive and analogical inferences are used. We do not detail in this paper each step of our system processing but we want to show how our work is related to abductive techniques, why it is faced to some similar problems and how solutions proposed for abduction can be adapted for EBL.

Keywords

Explanation-Based Learning Abduction Analogy 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Ayel B.E., Marquis P., Rusinowitch M., "Deductive/Abductive Diagnosis: The DA-Principles", Proceedings of ECAI 1990. pp.47–52.Google Scholar
  2. Cox P.T., Pietrzykowski T., "Causes for Events: their computation and applications", Proceedings of the 8th Conference on Automated Deduction, pp.608–621. Oxford, July 1986.Google Scholar
  3. DeJong G., Mooney R., "Explanation-Based Learning: An Alternative View", in Machine Learning 1, pp. 145–176. Kluwer Academic Publishers. 1986.Google Scholar
  4. DeJong G., "Plausible Inference vs. Abduction", Proceedings of AAAI Spring Symposium on Automated Abduction. pp.48–51. Standford University. March 1990.Google Scholar
  5. Duval B., "Abduction guidée par les Analogies entre Explications", Revue d'Intelligence Artificielle. Vol.4 No2. pp.11–27. Eds Hermès. 1990.Google Scholar
  6. Genest J., Matwin S.,Plante B., "Explanation-Based Learning with Incomplete Theories: A Three-Step Approach", Proceedings of the International Workshop on Machine Learning. pp.286–294. Austin 1990.Google Scholar
  7. Josephson J.R., "On the "Logical Form" of Abduction", Proceedings of the AAAI Spring Symposium on Automated Abduction. Standford University. pp.140–144. March 1990.Google Scholar
  8. Kedar-Cabelli S.T., McCarty L.T., "Explanation-Based Generalization as Resolution Theorem Proving", Proceedings of the Fourth International Machine Learning Workshop. pp.383–389. Irvine 1987.Google Scholar
  9. Kodratoff Y., "Using Abductive Recovery of Failed Proofs for Problem Solving by Analogy", Proceedings of the International Workshop on Machine Learning. pp.295–303. Austin 1990.Google Scholar
  10. Lloyd J.W., Foundations of Logic Programming Springer-Verlag. 1984.Google Scholar
  11. Michalski R.S., Kodratoff Y., "Research in Machine learning: Recent Progress, Classification of Methods and Future Direction" in Machine Learning: An Artificial intelligence Approach, Volume III, Y. Kodratoff and R.S. Michalski (Eds.), Morgan Kaufmann, San Mateo, 1990, pp. 3–30.Google Scholar
  12. Mitchell T., Keller R., Kedar-Cabelli S.T., "Explanation-Based Generalization: A Unifying View", in Machine Learning 1, pp.47–80. Kluwer Academic Publishers. 1986.Google Scholar
  13. Muggeleton S., Buntine W., "Machine Invention of First Order Predicates by Inverting Resolution", Proceedings of 5th International Machine Learning Workshop, pp.287–292. Morgan Kaufmann. 1988.Google Scholar
  14. Peirce C.S., "Elements of Logic" in Collected Papers of Charles Sanders Peirce (1839–1914), C.H. Hartshone and P. Weiss (Eds), The Belknap Press Harvard University Press, Cambridghe, MA. 1965.Google Scholar
  15. Poole D., "Variables in Hypotheses" Proceedings of the Tenth International Joint Conference on Artificial Intelligence. pp.905–908. Milan. Los-Altos, Ca: Morgan Kaufmann. 1987.Google Scholar
  16. Sadri F., Kowalski R., "A Theorem-Proving Approach to Database Integrity", in Foundations of Deductive Databases and Logic Programming pp.313–362. Jack Minker (Ed). Morgan Kaufmann Publishers, Los Altos, CA. 1987.Google Scholar
  17. Segre A.M., "On the Operationality/Generality Trade-off in Explanation-Based Learning", Proceedings of the Tenth International Joint Conference on Artificial Intelligence. pp.242–248. Milan. Los-Altos, Ca: Morgan Kaufmann. 1987.Google Scholar
  18. Shapiro E., Algorithmic Program Debugging, MIT Press, Cambridge, London. 1983.Google Scholar
  19. Stickel M.E. "A Prolog-like Inference System for Computing Minimum-Cost Abductive Explanations in Natural-Language Interpretation", Proceedings of the International Computer Science Conference '88, Hong Kong. Decembre 1988.Google Scholar
  20. Stickel M.E., "A Method for Abductive Reasoning in Natural Language Interpretation", Proceedings of the AAAI Spring Symposium on Automated Abduction. pp.5–9. Standford University. March 1990.Google Scholar
  21. Wirth R., "Learning by Failing to Prove" Proceedings of European Working Session on Learning, pp.237–251. Pitman. 1988.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • Béatrice Duval
    • 1
    • 2
  1. 1.Equipe Inférence et ApprentissageUniversité Paris SudOrsay CedexFrance
  2. 2.Université d'AngersAngers CedexFrance

Personalised recommendations