Research map / Learning theory and algorithms / Fairness
Too Relaxed to Be Fair
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 linear relaxation of fairness constraint (Zafar/Donini) with complex similarity-based classifiers (rbf kernel) | Zafar and Donini linear relaxations · Adult and Dutch | empirical DDP | Hurt |
| Replace linear relaxation with SearchFair (convex relaxation plus binary search on lambda for DDP=0) | Fair classifier training · Adult, Dutch, CelebA | DDP and DEO on test set | Helped |
| Use the Cotter et al. non-convex constrained method instead of SearchFair | Fair classifier training · Adult, Dutch, CelebA | DDP and DEO | No clear effect |
| Choose hyperparameters by most fair classifier under accuracy threshold (NVP) or accuracy under fairness threshold instead of accuracy-based CV | Baselines (Zafar, Donini, Cotter) · Multiple datasets | fairness scores | Helped |
Closest papers by meaning
- A Reductions Approach to Fair Classification (ICML 2018 · similarity 0.90)
- Empirical Risk Minimization Under Fairness Constraints (NeurIPS 2018 · similarity 0.89)
- Scalable and Stable Surrogates for Flexible Classifiers with Fairness Constraints (NeurIPS 2021 · similarity 0.89)
- Non-Discriminatory Machine Learning Through Convex Fairness Criteria (AAAI 2018 · similarity 0.89)
- Learning with Statistical Equality Constraints (NeurIPS 2025 · similarity 0.88)
- Individual Fairness In Strategic Classification (NeurIPS 2025 · 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.