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

As AI-based systems increasingly impact many areas of our lives, auditingthese systems for fairness is an increasingly high-stakes problem. Traditionalgroup fairness metrics can miss discrimination against individuals and aredifficult to apply after deployment. Counterfactual fairness describes anindividualized notion of fairness but is even more challenging to evaluateafter deployment. We present prediction sensitivity, an approach for continualaudit of counterfactual fairness in deployed classifiers. Predictionsensitivity helps answer the question: would this prediction have beendifferent, if this individual had belonged to a different demographic group --for every prediction made by the deployed model. Prediction sensitivity canleverage correlations between protected status and other features and does notrequire protected status information at prediction time. Our empirical resultsdemonstrate that prediction sensitivity is effective for detecting violationsof counterfactual fairness.

Prediction Sensitivity: Continual Audit of Counterfactual Fairness in Deployed Classifiers