AI Memo AIM-2003-019 Permutation Tests for Classification


We introduce and explore an approach to estimating statistical significance of classification accuracy, which is particularly useful in scientific applications of machine learning where high dimensionality of the data and the small number of training examples render most standard convergence bounds too loose to yield a meaningful guarantee of the… (More)


4 Figures and Tables

Cite this paper

@inproceedings{Mukherjee2003AIMA, title={AI Memo AIM-2003-019 Permutation Tests for Classification}, author={Sayan Mukherjee and Polina Golland and Dmitry Panchenko and Pablo Tamayo and Vladimir Koltchinskii and Jill P. Mesirov}, year={2003} }