Encyclopedia of Machine Learning

2010 Edition
| Editors: Claude Sammut, Geoffrey I. Webb

Epsilon Covers

  • Thomas Zeugmann
Reference work entry
DOI: https://doi.org/10.1007/978-0-387-30164-8_256


Let (M, ρ) be a metric space, let SM, and let ε > 0. A set EM is an ε-cover for S, if for every sS there is an eE such that ϱ(s, e) ≤ ε.

An ε-coverE is said to be proper, if ES.


The notion of an ε-cover is frequently used in kernel-based learning methods.

For further information, we refer the reader to Herbrich (2002).

Cross References

Recommended Reading

  1. Herbrich, R. (2002). Learning kernel classifiers: Theory and algorithms. Cambridge, MA: MIT Press.Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • Thomas Zeugmann

There are no affiliations available