Abstract
This chapter deals with discrimination rules that are picked from a certain class of classifiers by minimizing the empirical probability of error over a finite set of carefully selected rules. We begin with a class F of regression functions (i.e., a posteriori probability functions) η : R d → [0, 1] from which η n will be picked by the data. The massiveness of F can be measured in many ways—the route followed here is suggested in the work of Kolmogorov and Tikhomirov (1961). We will depart from their work only in details. We suggest comparing the results here with those from Chapters 12 and 15.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1996 Springer Science+Business Media New York
About this chapter
Cite this chapter
Devroye, L., Györfi, L., Lugosi, G. (1996). Epsilon Entropy and Totally Bounded Sets. In: A Probabilistic Theory of Pattern Recognition. Stochastic Modelling and Applied Probability, vol 31. Springer, New York, NY. https://doi.org/10.1007/978-1-4612-0711-5_28
Download citation
DOI: https://doi.org/10.1007/978-1-4612-0711-5_28
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4612-6877-2
Online ISBN: 978-1-4612-0711-5
eBook Packages: Springer Book Archive