ML Anthology

An Improved Predictive Accuracy Bound for Averaging Classifiers

John Langford, Matthias W. Seeger, Nimrod Megiddo
ICML 2001 pp. 290-297
/icml/2001/langford2001icml-improved/

Abstract

We present an improved bound on the difference between training and test errors for voting classiers. This improved averaging bound provides a theoretical justication for popular averaging techniques such as Bayesian classication, Maximum Entropy discrimination, Winnow and Bayes point machines and has implications for learning algorithm design. 1.

Semantic Scholar

Cite

Text

Langford et al. "An Improved Predictive Accuracy Bound for Averaging Classifiers." International Conference on Machine Learning, 2001.

Markdown

[Langford et al. "An Improved Predictive Accuracy Bound for Averaging Classifiers." International Conference on Machine Learning, 2001.](https://mlanthology.org/icml/2001/langford2001icml-improved/)

BibTeX

@inproceedings{langford2001icml-improved,
  title     = {{An Improved Predictive Accuracy Bound for Averaging Classifiers}},
  author    = {Langford, John and Seeger, Matthias W. and Megiddo, Nimrod},
  booktitle = {International Conference on Machine Learning},
  year      = {2001},
  pages     = {290-297},
  url       = {https://mlanthology.org/icml/2001/langford2001icml-improved/}
}