Abstract
Purchase PDF (1966 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (31)
Purchase PDF (2223 K)
)ln(1/δ)+VCdim(
doi:10.1016/0890-5401(91)90042-Z 
Copyright © 1991 Published by Elsevier Inc.
Equivalence of models for polynomial learnability*1
References and further reading may be available for this article. To view references and further reading you must purchase this article.
David Haussler, Michael Kearns† and Nick Littlestone†Manfred K. Warmuth
Received 24 January 1989;
Abstract
In this paper we consider several variants of Valiant's learnability model that have appeared in the literature. We give conditions under which these models are equivalent in terms of the polynomially learnable concept classes they define. These equivalences allow comparisons of most of the existing theorems in Valiant-style learnability and show that several simplifying assumptions on polynomial learning algorithms can be made without loss of generality. We also give a useful reduction of learning problems to the problem of finding consistent hypotheses, and give comparisons and equivalences between Valiant's model and the prediction learning models of Haussler, Littlestone, and Warmuth (in “29th Annual IEEE Symposium on Foundations of Computer Science,” 1988).
Article Outline
*1 The authors gratefully acknowledge the support of ONR Grant N00014-86-K-0454. M. Kearns was also supported by an AT & T Bell Laboratories Scholarship. This research was done while M. Kearns was visiting the University of California at Santa Cruz.
† Current address: AT & T Bell Laboratories, Murray Hill, New Jersey 07974.
