Research map / Learning theory and algorithms / Fairness
Two Simple Ways to Learn Individual Fairness Metrics from Data
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 a learned EXPLORE fair metric with SenSR rather than SenSR alone | SenSR on Adult · Adult | gender consistency (GR-Con.): .984 -> 0.987 | Helped |
| Use a learned EXPLORE fair metric with SenSR rather than SenSR alone | SenSR on Adult · Adult | Gap_G^RMS: .068 -> 0.065 | Helped |
| Exclude gender during classification (ablation) | SenSR+Explore on Adult · Adult | GR-Con.: 0.987 -> 0.993 | Helped |
| Exclude gender during classification (ablation) | SenSR+Explore on Adult · Adult | B-Acc,%: 79.4 -> 78.9 | Hurt |
| Exclude gender during classification (ablation) | SenSR+Explore on Adult · Adult | S-Con.: 0.966 -> 0.933 | Hurt |
| Learn and use a fair metric instead of Euclidean distance | Word embeddings · WEAT EAvAA / Pleasant vs. unpleasant | P: 0.00 -> 1e-2 | Helped |
| Learn and use a fair metric instead of Euclidean distance | Word embeddings · WEAT MNvFN / Career vs. family | P: 0.00 -> 2e-1 | Helped |
| Use a fair metric learned from baby names | Word embeddings with FACE-50 · WEAT MTHvART / Male vs. female terms | P: 8e-5 -> 6e-3 | No clear effect |
| Use 50 FACE factors rather than the Euclidean metric | Word embeddings · WEAT EAvAA / Pleasant vs. unpleasant | d: 1.36 -> 0.33 | Helped |
| Use 50 FACE factors rather than the Euclidean metric | Word embeddings · WEAT MTHvART / Male vs. female terms | d: 1.5 -> 1.16 | Helped |
Closest papers by meaning
- Average Individual Fairness: Algorithms, Generalization and Experiments (NeurIPS 2019 · similarity 0.91)
- Accurate Fairness: Improving Individual Fairness without Trading Accuracy (AAAI 2023 · similarity 0.90)
- Reconciling Predictive and Statistical Parity: A Causal Approach (AAAI 2024 · similarity 0.89)
- Individual Fairness In Strategic Classification (NeurIPS 2025 · similarity 0.89)
- The Fragility of Fairness: Causal Sensitivity Analysis for Fair Machine Learning (NeurIPS 2024 · similarity 0.89)
- Fairness without Harm: An Influence-Guided Active Sampling Approach (NeurIPS 2024 · similarity 0.88)
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.