Download PDF

Machine Learning

, Volume 11, Issue 2–3, pp 153–172 | Cite as

Multistrategy learning and theory revision

  • Lorenza Saitta
  • Marco Botta
  • Filippo Neri
Article
Received:
Accepted:

Abstract

This article presents the system WHY, which learns and updates a diagnostic knowledge base using domain knowledge and a set of examples. The a priori knowledge consists of a causal model of the domain that states the relationships among basic phenomena, and a body of phenomenological theory that describes the links between abstract concepts and their possible manifestations in the world. The phenomenological knowledge is used deductively, the causal model is used abductively, and the examples are used inductively. The problems of imperfection and intractability of the theory are handled by allowing the system to make assumptions during its reasoning. In this way, robust knowledge can be learned with limited complexity and a small number of examples. The system works in a first-order logic environment and has been applied in a real domain.

Keywords

Multistrategy learning causal models abduction diagnostic expert systems 
Download to read the full article text

References

  1. Baroglio, C., Botta, M. & Saitta, L. (1992). WHY: A system that learns using causal models and examples. In R. Michalski & G. Tecuci (Eds.), Machine learning: A multistrategy approach. Morgan Kaufmann.Google Scholar
  2. Bergadano, F., Giordana, A. & Saitta, L. (1988). Automated concept acquisition in noisy environment.IEEE Transactions PAMI, PAMI-10, 555–575.Google Scholar
  3. Bergadano, F. & Giordana, A. (1988). A knowledge intensive approach to concept induction.Proceedings of the Machine Learning Conference (pp. 305–317). Ann Arbor, MI.Google Scholar
  4. Bergadano, F., Giordana, A. & Ponsero, S. (1989). Deduction in top-down inductive learning.Proceedings of the Machine Learning Conference (pp. 23–25). Ithaca, NY.Google Scholar
  5. Bergadano, F., Giordana, A. & Saitta, L. (1990). Automated versus manual knowledge acquisition: a comparison in a real domain.Proceedings of the First Japanese Knowledge Acquisition for Knowledge-Based Systems Workshop (pp. 301–314). Tokyo, Japan.Google Scholar
  6. Botta, M. & Saitta, L. (1988). Improving knowledge base system performances by experience.Proceedings of the EWSL-88 (pp. 15–23). Glasgow, UK.Google Scholar
  7. Botta, M., Giordana, A. & Saitta, L. (1990). Knowledge base refinement using a causal model. In Z. Ras & M. Zemankova (Eds.),Intelligent systems: State of the art and future trends, Ellis-Horwood, Chichester, U.K.Google Scholar
  8. Cestnik, B. & Bratko, I. (1988). Learning redundant rules in noisy domains.Proceedings of the ECAI-88 (pp. 348–350). Munich, Germany.Google Scholar
  9. Chandrasekaran, B. & Mittal, S. (1983). Deep versus compiled knowledge approaches to diagnostic problem-solving.International Journal of Man-Machine Studies, 19, 425–436.Google Scholar
  10. Clark, K. (1978). Negation as failure. In H. Gallaire & J. Minker (Eds.),Logic and data bases. Plenum Press, New York, NY.Google Scholar
  11. Console, L., Torasso, P. & Theseider Dupré, D. (1990). A completion semantics for object-level abduction.Proceedings of the AAAI Symposium on Automated Abduction (pp. 72–75). Stanford, CA.Google Scholar
  12. Cox, P.T. & Pictrzykowski, T. (1987). General diagnosis by abductive inference.Proceedings of the IEEE Symposium on Logic Programming (pp. 183–189).Google Scholar
  13. Danyluk, A. (1991). Gemini: An integration of analytical and empirical learning.Proceedings of the First International Workshop on Multistrategy Learning (pp. 191–206). Harpers Ferry, WV.Google Scholar
  14. Davis, R. (1984). Diagnostic reasoning based on structure and behavior.Artificial Intelligence, 24, 347–410.Google Scholar
  15. DeJong, G. & Mooney, R. (1986). Explanation based learning: An alternative view.Machine Learning, 1, 47–80.Google Scholar
  16. DeJong, G. (1990). Plausible inference vs. abduction.Proceedings of the AAAI Symposium on Automated Abduction (pp. 48–51). Stanford, CA.Google Scholar
  17. de Kleer, J. & Seely Brown, J. (1986). Theories of causal ordering.Artificial Intelligence, 29, 33–61.Google Scholar
  18. De Raedt, L. & Bruynooghe, M. (1991). CLINT: a multistrategy interactive concept learner and theory revision system.Proceedings of the First International Workshop on Multistrategy Learning (pp. 175–190). Harpers Ferry, WV.Google Scholar
  19. Genesereth, M. (1984). The use of design descriptions in automated diagnosis.Artificial Intelligence, 24, 411–436.Google Scholar
  20. Giordana, A., Saitta, L., Bergadano, F., Brancadori, F. & DeMarchi, D. (1993). ENIGMA: A system that learns diagnostic knowledge.IEEE Trans. on Knowledge and Data Engineering, 5 (1).Google Scholar
  21. Hirsh, H. (1988). Reasoning about operationality for explanation-based learning.Proceedings of the Machine Learning Conference (pp. 214–220). Ann Arbor, MI.Google Scholar
  22. Kahn, G. (1984). On when diagnostic systems want to do without causal knowledge.Advances in Artificial Intelligence, (pp. 21–30).Google Scholar
  23. Keller, R. (1988). Defining operationality for EBL.Artificial Intelligence, 35, 227–242.Google Scholar
  24. Kodratoff, Y. (1991). Induction and the organization of knowledge.Proceedings of the First International Workshop on Multistrategy Learning (pp. 34–48). Harpers Ferry, WV. (See also this issue.)Google Scholar
  25. Lebowitz, M. (1986). Integrated learning: Controlling explanation.Cognitive Science, 10, 219–240.Google Scholar
  26. Matwin, S. & Plante, B. (1991). A deductive-inductive method for theory revision.Proceedings of the International Workshop on Multistrategy Learning (pp. 160–174). Harpers Ferry, WV. (See also this issue.)Google Scholar
  27. Michalski, R. (1983). A theory and methodology of inductive learning.Artificial Intelligence, 20, 111–161.Google Scholar
  28. Michalski, R. (1991). Inferential learning theory as a basis for multistrategy task-adaptive learning.Proceedings of the First International Workshop on Multistrategy Learning (pp. 3–18). Harpers Ferry, WV.Google Scholar
  29. Michalski, R.S. & Chilausky, R.L. (1991). Learning by being told and learning from examples: an experimental comparison of the two methods of knowledge acquisition in the context of developing an expert system for soybean disease diagnosis.International Journal of Policy Analysis and Information Systems, 4, 125–126.Google Scholar
  30. Mitchell, T. (1982). Generalization as search.Artificial Intelligence, 18, 203–226.Google Scholar
  31. Mitchell, T., Keller, R. & Kedar-Cabelli, S. (1986). Explanation based generalization.Machine Learning, 1, 47–80.Google Scholar
  32. Mooney, R.J. & Ourston, D. (1991). A multistrategy approach to theory refinement.Proceedings of the First International Workshop on Multistrategy Learning (pp. 115–131). Harpers Ferry, WV.Google Scholar
  33. Morris, S. & O'Rorke, P. (1990). An approach to theory revision using abduction.Proceedings of the AAAI Symposium on Automated Abduction (pp. 33–37). Standford, CA.Google Scholar
  34. Pazzani, M.J. (1988). Integrating explanation-based and empirical learning methods in OCCAM.Proceedings of the European Working Session on Learning (pp. 147–165). Glasgow, UK.Google Scholar
  35. Poole, D. (1988. Representing knowledge for logic-based diagnosis.Proceedings of the International Conference on Fifth Generation Computer Systems (pp. 1282–1290). Tokyo, Japan.Google Scholar
  36. Quinlan, R. (1986). Induction of decision trees.Machine Learning, 1, 81–106.Google Scholar
  37. Reiter, R. (1984). A theory of diagnosis from first principles.Artificial Intelligence, 32, 57–95.Google Scholar
  38. Robinson, J.A. & Siebert, E.E. (1982). LOGLISP: An alternative to Prolog.Machine Intelligence, 10, 399–419.Google Scholar
  39. Saitta, L., Botta, M., Ravotto, S. & Sperotto, S.B. (1991). Improving learning by using deep models.Proceedings of the First International Workshop on Multistrategy Learning (pp. 131–143). Harpers Ferry, WV.Google Scholar
  40. Segre, A.M. (1987). On the operationality/generality trade-off in explanation-based learning.Proceedings of the IJCAI-87 (pp. 242–248). Milan, Italy.Google Scholar
  41. Tecuci, G. (1991). Learning as understanding the external world.Proceedings of the First International Workshop on Multistrategy Learning (pp. 49–64). Harpers Ferry, WV.Google Scholar
  42. Torasso, P. & Console, L. (1989).Diagnostic problem solving. Van Nostrand Reinhold, New York, NY.Google Scholar

Copyright information

© Kluwer Academic Publishers 1993

Authors and Affiliations

  • Lorenza Saitta
    • 1
  • Marco Botta
    • 1
  • Filippo Neri
    • 1
  1. 1.Dipartimento di InformaticaUniversità di TorinoTorinoItaly

Personalised recommendations