Abstract
Universal consistency gives us a partial satisfaction—without knowing the underlying distribution, taking more samples is guaranteed to push us close to the Bayes rule in the long run. Unfortunately, we will never know just how close we are to the Bayes rule unless we are given more information about the unknown distribution (see Chapter 7). A more modest goal is to do as well as possible within a given class of rules.
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). Splitting the Data. 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_22
Download citation
DOI: https://doi.org/10.1007/978-1-4612-0711-5_22
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