English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We revisit resampling procedures for error estimation in binaryclassification in terms of U-statistics. In particular, we exploit the factthat the error rate estimator involving all learning-testing splits is aU-statistic. Thus, it has minimal variance among all unbiased estimators and isasymptotically normally distributed. Moreover, there is an unbiased estimatorfor this minimal variance if the total sample size is at least the doublelearning set size plus two. In this case, we exhibit such an estimator which isanother U-statistic. It enjoys, again, various optimality properties and yieldsan asymptotically exact hypothesis test of the equality of error rates when twolearning algorithms are compared. Our statements apply to any deterministiclearning algorithms under weak non-degeneracy assumptions.

A U-statistic estimator for the variance of resampling-based error estimators