Research map / Learning theory and algorithms / Fairness
Bipartite Ranking Fairness Through a Model Agnostic Ordering Adjustment
Sen Cui, Weishen Pan, Changshui Zhang, Fei Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| adjust cross-group ordering with xOrder instead of score transformation (post-logit) | linear model post-processing · Adult | xAUC(a,b): 0.879 -> 0.896 | Improved |
| apply post-logit score transformation | unadjusted linear model · Adult | ΔxAUC: 0.11 -> 0.02 | Improved |
| apply post-logit score transformation | unadjusted linear model · COMPAS | ΔxAUC: 0.23 -> 0.03 | Improved |
| use xOrder instead of post-logit when train and test score distributions differ (RankBoost) | post-processing · Adult and COMPAS (RankBoost) | ΔxAUC | Improved |
| reduce training samples to 5000 for xOrder versus baselines | xOrder · Adult (5000 training samples) | ΔxAUC: 0.02 -> 0.01 | Improved |
| learn xOrder adjustment from an equidistant subset of 2000 samples (ablation) | xOrder · large eICU (prolonged LOS, race) | fairness-utility trade-off | No clear effect |
| use iterative xOrder for 5 and 10 protected groups | xOrder · large eICU (mortality, age groups) | disparity | Improved |
| use xOrder instead of corr-reg and opti-tfco | linear model · COMPAS and MIMIC-III | ΔxAUC | Improved |
Closest papers by meaning
- Stability and Multigroup Fairness in Ranking with Uncertain Predictions (ICML 2024 · similarity 0.88)
- The Fairness of Risk Scores Beyond Classification: Bipartite Ranking and the XAUC Metric (NeurIPS 2019 · similarity 0.87)
- Policy Learning for Fairness in Ranking (NeurIPS 2019 · similarity 0.87)
- Post-hoc bias scoring is optimal for fair classification (ICLR 2024 · similarity 0.84)
- Bipartite Ranking From Multiple Labels: On Loss Versus Label Aggregation (ICML 2025 · similarity 0.84)
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
