Research map / Learning theory and algorithms / Fairness
Two-sided fairness in rankings via Lorenz dominance
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 |
|---|---|---|---|
| decrease welfare curvature alpha from 1 to -5 (focus on worse-off users) | welf · Higgs Twitter reciprocal recommendation | cumulative utility of 10% worse-off users: 120 vs 280 | Helped |
| decrease welfare curvature alpha from 1 to -5 | welf · Higgs Twitter reciprocal recommendation | total utility (thousands): 17 vs 6.4 | Hurt |
| decrease user curvature alpha_1 from 1 to -2 | welf · Lastfm-2k one-sided recommendation | user inequality vs total user utility | Mixed |
| welf alpha_1=0 instead of Patro et al. | two-sided fair ranking · Lastfm-2k | total user utility | Helped |
| welf instead of quality-weighted exposure | item-side fair ranking · Lastfm-2k | user utility vs item inequality | Helped |
| welf instead of Patro et al., eq. exposure, qua.-weighted | item-side fair ranking · Lastfm-2k, Gini | total user utility | Helped |
| measure item inequality with standard deviation instead of Gini | comparison welf vs eq. exposure · Lastfm-2k | user utility vs item inequality | Hurt |
| welf instead of eq. utility | reciprocal recommendation · Higgs Twitter | utility of worse-off users at given total utility | Helped |
| welf instead of quality-weighted and eq. exposure | reciprocal recommendation · Higgs Twitter | utility trade-off | Helped |
Closest papers by meaning
- User-item fairness tradeoffs in recommendations (NeurIPS 2024 · similarity 0.90)
- Interpolating Item and User Fairness in Multi-Sided Recommendations (NeurIPS 2024 · similarity 0.88)
- Fair Ranking with Noisy Protected Attributes (NeurIPS 2022 · similarity 0.88)
- Optimizing Long-term Social Welfare in Recommender Systems: A Constrained Matching Approach (ICML 2020 · similarity 0.88)
- Fairly Recommending with Social Attributes: A Flexible and Controllable Optimization Approach (NeurIPS 2023 · similarity 0.87)
- Fairness in Aggregation: Optimal Top-$k$ and Improved Full Ranking (ICML 2026 · 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.