Advertisement

Combinatorial results on the complexity of teaching and learning

  • Tibor Hegedüs
Contributions
Part of the Lecture Notes in Computer Science book series (LNCS, volume 841)

Abstract

Some recent work in computational learning theory has focused on the complexity of teaching by examples. In this paper we study two combinatorial measures expressing the number of examples needed to teach any consistent learner, called teaching dimension and universal teaching dimension. We give a general lower bound on the teaching dimension that improves previous results, relate the teaching complexity measures to some learning complexity measures and combinatorial parameters, and compute bounds on the teaching dimension(s) of natural Boolean concept classes. We also observe an analogy between the teaching model and some problems in the fault detection research, and make use of some results achieved within that framework.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    D. Angluin, “Queries and Concept Learning”, Machine Learning 2 (1988) 319–342.Google Scholar
  2. 2.
    M. Anthony, G. Brightwell, D. Cohen and J. Shawe-Taylor, “On Exact Specification by Examples”, in: Proceedings of the 5th Annual ACM Workshop on Computational Learning Theory (COLT'92), ACM Press, New York, NY, 1992, pp. 311–318.Google Scholar
  3. 3.
    A. Blum, P. Chalasani and J. Jackson, “On Learning Embedded Symmetric Concepts” in: Proceedings of the 6th Annual ACM Conference on Computational Learning Theory (COLT'93), ACM Press, New York, NY, 1993, pp. 337–346.Google Scholar
  4. 4.
    M. A. Breuer and A. D. Friedman, Diagnosis & Reliable Design of Digital Systems, Computer Science Press, Rockville, MD, 1976.Google Scholar
  5. 5.
    N. H. Bshouty and R. Cleve, “On the Exact Learning of Formulas in Parallel”, in: Proceedings of the 33rd Annual Symposium on Foundations of Computer Science (FOCS'92), IEEE Computer Society Press, Los Alamitos, CA, 1992, pp. 513–522.Google Scholar
  6. 6.
    I. A. Chegis and S. V. Yablonskii, “Logical Methods of Control of the Work of Electrical Circuits”, Trudy Matematicheskogo Instituta Akad. Nauk SSSR Imeni V. A. Steklova 51 (1958) 270–360 (in Russian).Google Scholar
  7. 7.
    A. D. Friedman and P. R. Menon, Fault Detection in Digital Circuits, Prentice-Hall, Englewood Cliffs, NJ, 1971.Google Scholar
  8. 8.
    S. A. Goldman and M. J. Kearns, “On the Complexity of Teaching”, in: Proceedings of the 4th Annual Workshop on Computational Learning Theory (COLT'91), Morgan Kaufmann, San Mateo, CA, 1991, pp. 303–314.Google Scholar
  9. 9.
    S. A. Goldman, M. J. Kearns, and R. E. Schapire, “Exact Identification of Read-Once Formulas Using Fixed Points of Amplification Functions”, SIAM Journal on Computing 22(4) (1993) 705–726.Google Scholar
  10. 10.
    S. A. Goldman and H. D. Mathias, “Teaching a Smarter Learner”, in: Proceedings of the 6th Annual ACM Conference on Computational Learning Theory (COLT'93), ACM Press, New York, NY, 1993, pp. 67–76.Google Scholar
  11. 11.
    T. Hegedüs, “On Training Simple Neural Networks and Small-Weight Neurons”, in: Proceedings of the 1st European Conference on Computational Learning Theory (EuroCOLT'93), Royal Holloway, University of London, December 1993, Oxford University Press, to appear.Google Scholar
  12. 12.
    J. Jackson and A. Tomkins, “A Computational Model of Teaching”, in: Proceedings of the 5th Annual ACM Workshop on Computational Learning Theory (COLT'92), ACM Press, New York, NY, 1992, pp. 319–326.Google Scholar
  13. 13.
    W. Maass and Gy. Turán, “Lower Bound Methods and Separation Results for On-Line Learning Models”, Machine Learning 9 (1992) 107–145.Google Scholar
  14. 14.
    A. Shinohara and S. Miyano, “Teachability in Computational Learning”, New Generation Computing 2 (1991) 337–347.Google Scholar
  15. 15.
    N. A. Solovev, Tests (theory, design, application), Nauka, Novosibirsk, 1978 (in Russian).Google Scholar
  16. 16.
    S. V. Yablonskii, “Test in Cybernetics”, in: Mathematical Encyclopedia, vol. 5 [SluYa], (I. M. Vinogradov, ed.), Sovet. Entsiklopediya, Moscow, 1985, pp. 342–346 (in Russian).Google Scholar
  17. 17.
    S. V. Yablonskii and I. A. Chegis, “On Tests for Electrical Circuits”, Uspekhi Matematicheskikh Nauk 10(4) (1955) 182–184 (in Russian).Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1994

Authors and Affiliations

  • Tibor Hegedüs
    • 1
  1. 1.Department of Computer ScienceComenius UniversityBratislavaSlovakia

Personalised recommendations