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 propose the holdout randomization test (HRT), an approach to featureselection using black box predictive models. The HRT is a specialized versionof the conditional randomization test (CRT; Candes et al., 2018) that uses datasplitting for feasible computation. The HRT works with any predictive model andproduces a valid $p$-value for each feature. To make the HRT more practical, wepropose a set of extensions to maximize power and speed up computation. Insimulations, these extensions lead to greater power than a competingknockoffs-based approach, without sacrificing control of the error rate. Weapply the HRT to two case studies from the scientific literature whereheuristics were originally used to select important features for predictivemodels. The results illustrate how such heuristics can be misleading relativeto principled methods like the HRT. Code is available athttps://github.com/tansey/hrt.

The Holdout Randomization Test for Feature Selection in Black Box Models