Research map / Learning theory and algorithms / Fairness
On preserving non-discrimination when combining expert advice
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 separate multiplicative weights instances for each group instead of one shared instance | Online expert advice with individually equalized-error-rate-fair experts · Equalized error rates | fairness gap: α-fairness | Helped |
| Use separate multiplicative weights instances for each group | Online expert advice · Equalized error rates | ε-approximate regret: O(|G| log(d)/ε) | Helped |
| Combine groups with a single group-unaware algorithm | Online expert advice with individually fair experts · Equalized error rates | fairness gap: α-unfair for any α>0 | Hurt |
| Use group-unaware no-regret learning | Online expert advice with individually equalized-odds-fair predictors · False negative rate | fairness gap: 3/8 | Hurt |
| Use group-aware no-regret learning | Online expert advice with individually equalized-odds-fair predictors · False negative rate | fairness gap: (0.49-0.99b)/(1-b) | Hurt |
| Use shifting-regret algorithms, even when run separately for each group | Online expert advice · Equalized error rates | fairness gap: α-unfair for any α<1/2 | Hurt |
Closest papers by meaning
- Accuracy vs. Accuracy: Computational Tradeoffs Between Classification Rates and Utility (NeurIPS 2026 · similarity 0.87)
- Monotone Individual Fairness (ICML 2024 · similarity 0.86)
- The Stability of Online Algorithms in Performative Prediction (NeurIPS 2026 · similarity 0.85)
- Remembering to Be Fair: Non-Markovian Fairness in Sequential Decision Making (ICML 2024 · similarity 0.85)
- Multigroup Fairness and Omniprediction: Separations and Equivalences (NeurIPS 2026 · similarity 0.85)
- A Comprehensive View of Fairness through Distributional Stability (NeurIPS 2026 · similarity 0.85)
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.