<ul><li>Fairness-aware learning aims to mitigate discrimination against specific protected social groups.</li><li>Training and test data sampling can affect the reliability of reported fairness metrics.</li><li>FairMatch, a post-processing method, utilizes propensity score matching to evaluate and mitigate bias.</li><li>Experimental results show that FairMatch improves fairness evaluation and mitigation without sacrificing predictive performance.</li></ul>

Whence Is A Model Fair? Fixing Fairness Bugs via Propensity Score Matching

Discover more