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

The Right to Explanation is an important regulatory principle that allowsindividuals to request actionable explanations for algorithmic decisions.However, several technical challenges arise when providing such actionableexplanations in practice. For instance, models are periodically retrained tohandle dataset shifts. This process may invalidate some of the previouslyprescribed explanations, thus rendering them unactionable. But, it is unclearif and when such invalidations occur, and what factors determine explanationstability i.e., if an explanation remains unchanged amidst model retraining dueto dataset shifts. In this paper, we address the aforementioned gaps andprovide one of the first theoretical and empirical characterizations of thefactors influencing explanation stability. To this end, we conduct rigoroustheoretical analysis to demonstrate that model curvature, weight decayparameters while training, and the magnitude of the dataset shift are keyfactors that determine the extent of explanation (in)stability. Extensiveexperimentation with real-world datasets not only validates our theoreticalresults, but also demonstrates that the aforementioned factors dramaticallyimpact the stability of explanations produced by various state-of-the-artmethods.

On Minimizing the Impact of Dataset Shifts on Actionable Explanations