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/

Zendy Plus

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

Zendy Tools

Zendy Open

Bipartite ranking, which aims to learn a scoring function that ranks positiveindividuals higher than negative ones from labeled data, is widely adopted invarious applications where sample prioritization is needed. Recently, therehave been rising concerns on whether the learned scoring function can causesystematic disparity across different protected groups defined by sensitiveattributes. While there could be trade-off between fairness and performance, inthis paper we propose a model agnostic post-processing framework for balancingthem in the bipartite ranking scenario. Specifically, we maximize a weightedsum of the utility and fairness by directly adjusting the relative ordering ofsamples across groups. By formulating this problem as the identification of anoptimal warping path across different protected groups, we propose anon-parametric method to search for such an optimal path through a dynamicprogramming process. Our method is compatible with various classificationmodels and applicable to a variety of ranking fairness metrics. Comprehensiveexperiments on a suite of benchmark data sets and two real-world patientelectronic health record repositories show that our method can achieve a greatbalance between the algorithm utility and ranking fairness. Furthermore, weexperimentally verify the robustness of our method when faced with the fewertraining samples and the difference between training and testing ranking scoredistributions.

Towards Model-Agnostic Post-Hoc Adjustment for Balancing Ranking Fairness and Algorithm Utility