Research map / Learning theory and algorithms / Fairness
Fair Ranking with Noisy Protected Attributes
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use NResilient, which incorporates probabilistic protected-attribute information, rather than attribute-based fair-ranking baselines | Fair ranking on synthetic data · Synthetic data | RD: ≈0.81; MC ≈0.79; CSV, SJ, and GAK ≤0.68; Uncons ≈0.75 | Helped |
| Use NResilient instead of CSV, SJ, or GAK with inferred groups | Fair ranking on Occupations image data · Occupations image data | RD: ≈0.81 vs ≤0.77 | Helped |
| Use NResilient instead of MC | Fair ranking on Occupations image data · Occupations image data | RD: ≈0.81 vs ≤0.70 | Helped |
| Use NResilient instead of the other fair-ranking baselines | Fair ranking on name data with four groups · Name data, four groups | RD: ≈0.67 vs ≈0.61 for CSV, GAK, and SJ; ≤ 0.53 for MC | Helped |
| Account for probabilistic attribute noise instead of applying existing fair-ranking algorithms to inferred groups | Fair ranking with noisy attributes · Synthetic data, φ≤1.2 | RD: NResilient ≈0.81; CSV, SJ, and GAK ≤0.68; Uncons ≈0.75 | Helped |
| Use NResilient rather than the baselines at RD≥0.66 | Fair ranking on Occupations image data · Occupations image data | utility at a given RD | Helped |
| Use NResilient rather than CSV or SJ at selection-lift 0.55 | Fair ranking in the additional simulation · Additional simulation, selection-lift 0.55 | utility: 2% higher utility | Helped |
| Use NResilient instead of any baseline across tested noise levels | Fair ranking on Occupations image data · Occupations image data, varying noise | RD | Mixed |
Closest papers by meaning
- Pairwise Fairness for Ranking and Regression (AAAI 2020 · similarity 0.91)
- Fairness in Aggregation: Optimal Top-$k$ and Improved Full Ranking (ICML 2026 · similarity 0.90)
- On the Problem of Underranking in Group-Fair Ranking (ICML 2021 · similarity 0.89)
- Stability and Multigroup Fairness in Ranking with Uncertain Predictions (ICML 2024 · similarity 0.89)
- Individual Fairness In Strategic Classification (NeurIPS 2025 · similarity 0.87)
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.87)
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.