An Error Bound Based on a Worst Likely Assignment

Abstract

This paper introduces a new PAC transductive error bound for classification. The method uses information from the training examples and inputs of working examples to develop a set of likely assignments to outputs of the working examples. A likely assignment with maximum error determines the bound. The method is very effective for small data sets.

Cite

Text

Bax and Callejas. "An Error Bound Based on a Worst Likely Assignment." Journal of Machine Learning Research, 2008.

Markdown

[Bax and Callejas. "An Error Bound Based on a Worst Likely Assignment." Journal of Machine Learning Research, 2008.](https://mlanthology.org/jmlr/2008/bax2008jmlr-error/)

BibTeX

@article{bax2008jmlr-error,
  title     = {{An Error Bound Based on a Worst Likely Assignment}},
  author    = {Bax, Eric and Callejas, Augusto},
  journal   = {Journal of Machine Learning Research},
  year      = {2008},
  pages     = {859-891},
  volume    = {9},
  url       = {https://mlanthology.org/jmlr/2008/bax2008jmlr-error/}
}