Advertisement

Learning First-Order Definite Theories via Object-Based Queries

  • Joseph Selman
  • Alan Fern
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6913)

Abstract

We study the problem of exact learning of first-order definite theories via queries, toward the goal of allowing humans to more efficiently teach first-order concepts to computers. Prior work has shown that first order Horn theories can be learned using a polynomial number of membership and equivalence queries [6]. However, these query types are sometimes unnatural for humans to answer and only capture a small fraction of the information that a human teacher might be able to easily communicate. In this work, we enrich the types of information that can be provided by a human teacher and study the associated learning problem from a theoretical perspective. First, we consider allowing queries that ask the teacher for the relevant objects in a training example. Second, we examine a new query type, called a pairing query, where the teacher provides mappings between objects in two different examples. We present algorithms that leverage these new query types as well as restrictions applied to equivalence queries to significantly reduce or eliminate the required number of membership queries, while preserving polynomial learnability. In addition, we give learnability results for certain cases of imperfect teachers. These results show, in theory, the potential for incorporating object-based queries into first-order learning algorithms in order to reduce human teaching effort.

Keywords

Inductive Logic Programming Relevant Object Query Type Target Theory Membership Query 
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.

References

  1. 1.
    Angluin, D.: Queries and concept learning. Machine Learning 2, 319–342 (1988), http://dx.doi.org/10.1023/A:1022821128753, doi:10.1023/A:1022821128753MathSciNetGoogle Scholar
  2. 2.
    Angluin, D., Frazier, M., Pitt, L.: Learning conjunctions of horn clauses. Machine Learning 9, 147–164 (1992), http://dx.doi.org/10.1007/BF00992675 zbMATHGoogle Scholar
  3. 3.
    Arias, M., Khardon, R., Maloberti, J.: Learning horn expressions with LOGAN-H. The Journal of Machine Learning Research 8, 549–587 (2007)zbMATHGoogle Scholar
  4. 4.
    Baum, E.B., Lang, K.: Query learning can work poorly when a human oracle is used. In: IJCNN (1992)Google Scholar
  5. 5.
    Feldman, V., Shah, S.: Separating models of learning with faulty teachers. Theor. Comput. Sci. 410, 1903–1912 (2009), http://portal.acm.org/citation.cfm?id=1519541.1519713 CrossRefzbMATHMathSciNetGoogle Scholar
  6. 6.
    Khardon, R.: Learning function-free horn expressions. Machine Learning 37, 241–275 (1999), http://dx.doi.org/10.1023/A:1007610422992 CrossRefzbMATHGoogle Scholar
  7. 7.
    McDaniel, R.G., Myers, B.A.: Getting more out of programming-by-demonstration. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems: the CHI is the Limit, CHI 1999, pp. 442–449. ACM, New York (1999), http://doi.acm.org/10.1145/302979.303127 Google Scholar
  8. 8.
    Plotkin, G.D.: A Note on Inductive Generalization. Machine Intelligence 5, 153–163 (1970)zbMATHMathSciNetGoogle Scholar
  9. 9.
    Raedt, L.D., Dzeroski, S.: First-order jk-clausal theories are pac-learnable. Artificial Intelligence 70(1-2), 375–392 (1994), http://www.sciencedirect.com/science/article/B6TYF-47WT97K-F/2/55a8bf6608387d13b5bfa7a55e4b50d2 CrossRefzbMATHMathSciNetGoogle Scholar
  10. 10.
    Reddy, C., Tadepalli, P.: Learning first-order acyclic horn programs from entailment. In: Page, D. (ed.) ILP 1998. LNCS, vol. 1446, pp. 23–37. Springer, Heidelberg (1998), http://dx.doi.org/10.1007/BFb0027308 CrossRefGoogle Scholar
  11. 11.
    Reddy, C., Tadepalli, P.: Learning horn definitions: Theory and an application to planning. New Generation Computing 17, 77–98 (1999), http://dx.doi.org/10.1007/BF03037583, doi:10.1007/BF03037583CrossRefGoogle Scholar
  12. 12.
    Valiant, L.G.: A theory of the learnable. Commun. ACM 27, 1134–1142 (1984), http://doi.acm.org/10.1145/1968.1972 CrossRefzbMATHGoogle Scholar
  13. 13.
    Walker, T., Natarajan, S., Kunapuli, G., Shavlik, J., Page, D.: Automation of ilp setup and search via user provided relevance and type information. In: Inductive Logic Programming (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Joseph Selman
    • 1
  • Alan Fern
    • 1
  1. 1.School of Electrical Engineering and Computer ScienceOregon State UniversityUSA

Personalised recommendations