Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Individually Fair Learning with One-Sided Feedback
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 the reduction with Exp2 and C=T^\frac{1}{5} | Individually fair online learning with one-sided feedback · Adversarial online learning, comparator Q_{\alpha-\epsilon,\gamma} | accuracy regret: O\left(k^\frac{3}{2}T^\frac{4}{5}\log\vert\mathcal{H}\vert^\frac{1}{2}\right) | Helped |
| Use the reduction with Exp2 and C=T^\frac{1}{5} | Individually fair online learning with one-sided feedback · Adversarial online learning | fairness violations: O\left(\frac{1}{\epsilon}k^\frac{3}{2}T^\frac{4}{5}\log\vert\mathcal{H}\vert^\frac{1}{2}\right) | Helped |
| Use resampling to estimate the implicit policy before requesting panel feedback | Context-Semi-Bandit-FTPL · Adversarial online learning, comparator Q_{\alpha-\epsilon,\gamma} | accuracy regret: \tilde{O}\left(k^\frac{11}{4}s^\frac{3}{4}T^\frac{41}{45}\log\vert\mathcal{H}\vert^\frac{1}{2}\right) | Helped |
| Use resampling to estimate the implicit policy before requesting panel feedback | Context-Semi-Bandit-FTPL · Adversarial online learning | fairness violations: \tilde{O}\left(\frac{1}{\epsilon}k^\frac{11}{4}s^\frac{3}{4}T^\frac{41}{45}\log\vert\mathcal{H}\vert^\frac{1}{2}\right) | Helped |
| Query auditors on a sampled classifier rather than on the randomized policy | Randomized policy auditing · Appendix counterexample with k=2 | unfairness loss: 1 vs 0 | Hurt |
Closest papers by meaning
- Monotone Individual Fairness (ICML 2024 · similarity 0.93)
- Cost Efficient Fairness Audit Under Partial Feedback (NeurIPS 2026 · similarity 0.86)
- Fair Classification with Partial Feedback: An Exploration-Based Data Collection Approach (ICML 2024 · similarity 0.85)
- Bandit-Feedback Online Multiclass Classification: Variants and Tradeoffs (NeurIPS 2024 · similarity 0.85)
- Individual Fairness In Strategic Classification (NeurIPS 2025 · similarity 0.85)
- Fair Off-Policy Learning from Observational Data (ICML 2024 · 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.