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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Machine learning algorithms are extensively used to make increasingly moreconsequential decisions about people, so achieving optimal predictiveperformance can no longer be the only focus. A particularly importantconsideration is fairness with respect to race, gender, or any other sensitiveattribute. This paper studies intersectional fairness, where intersections ofmultiple sensitive attributes are considered. Prior research has mainly focusedon fairness with respect to a single sensitive attribute, with intersectionalfairness being comparatively less studied despite its critical importance forthe safety of modern machine learning systems. We present a comprehensiveframework for auditing and achieving intersectional fairness in classificationproblems: we define a suite of metrics to assess intersectional fairness in thedata or model outputs by extending known single-attribute fairness metrics, andpropose methods for robustly estimating them even when some intersectionalsubgroups are underrepresented. Furthermore, we develop post-processingtechniques to mitigate any detected intersectional bias in a classificationmodel. Our techniques do not rely on any assumptions regarding the underlyingmodel and preserve predictive performance at a guaranteed level of fairness.Finally, we give guidance on a practical implementation, showing how theproposed methods perform on a real-world dataset.

Auditing and Achieving Intersectional Fairness in Classification Problems