Abstract
In this paper the problem of using queries in order to learn an unknown concept is investigated. Unfortunately, for many concept classes an exhaustive or nearly exhaustive search is necessary when using membership queries only (so named by Angluin). Not only membership queries but also superset queries and equivalence queries as defined by Angluin are considered. Furthermore, learning protocols for which efficient query learning is possible are introduced. At first, the paper introduces a characterization of concept classes where exhaustive membership queries can be avoided by using clever query strategies. The Vapnik-Chervonenkis dimension for stating an upper-bound for required membership queries is used along with a sketch of an appropriate algorithm. In addition, concept classes are characterized where one or some more positive examples provided to the learning system allow to dramatically reduce the number of required queries. Furthermore, for these concept classes the existence of efficient query strategies which use combinations of different query types is proved. A probabilistic query learning algorithm is presented as well. Finally, concept classes are characterized for which a polynomial time learning algorithm exists which uses membership queries efficiently.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
D. Angluin: Queries and Concept Learning; Machine Learning (2), 1988, pp. 319–342
D. Angluin: Learning k-term DNF formulas using queries and counterexamples; Technical Report YALEU/DCS/RR-559). New Haven, CT: Yale University, Department of Computer Science
D. Angluin: Learning regular sets from queries and counterexamples; Information and Computation, 75,pp. 87–106
D. Angluin, L. Ilellerstein, M. Karpinski: Learning Read-Once Formulas with Queries, Report No. UCB/CSD 89/528, August 1989. Berkeley, CA: University of California CSD (EECS).
A. Blumer, A. Ehrenfeucht, D. Haussler, M. Warmuth: Classifying learnable geometric concepts with the Vapnik-Chervonenkis dimension; Proceedings 18th Symp. on Theory of Computing1986; pp. 273–282
W. Gasarch, C. Smith: Learning via Queries; Proceedings of the first Workshop on Computational Learning Theory MIT 1988, Los Altos, CA: Morgan Kaufmann Publishers 1988; pp. 227–241
L. Hellerstein: Learning read-once formulas using membership queries. Proceedings of the Second Annual Workshop on Computational Learning Theory, Los Altos, CA: Morgan Kaufmann Publishers 1989; pp. 146–161
A. Hoffmann: Unifying several learning situations; Proceedings of the first Workshop on Computational Learning Theory MIT 1988, Los Altos, CA: Morgan Kaufmann Publishers 1988; pp. 415–416
C. Sammut, R. Banerji: Learning concepts by asking questions. In R.S. Michalski, J.G. Carbonell, T.M. Mitchell (Eds.): Machine Learning: An Artificial Intelligence Approach (Vol.2). Los Altos, CA: Morgan Kaufmann Publishers 1986; pp. 167–191
L. Valiant: A theory of the learnable. Communications of the ACM, 27, pp. 1134–1142, 1984
V. N. Vapnik und A. Ya. Chervonenkis: On the uniform convergence of relative frequencies of events to their probabilities; Theory of Probability and its Applications 16(2), 1971; pp. 264–280
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1990 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Hoffmann, A.G. (1990). Types of Efficient Query Learning. In: Marburger, H. (eds) GWAI-90 14th German Workshop on Artificial Intelligence. Informatik-Fachberichte, vol 251. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-76071-6_29
Download citation
DOI: https://doi.org/10.1007/978-3-642-76071-6_29
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-53132-6
Online ISBN: 978-3-642-76071-6
eBook Packages: Springer Book Archive