Research map / Learning theory and algorithms / Combinatorial and search optimization
Statistical Guarantees in the Search for Less Discriminatory Algorithms
Chris Hays, Benjamin Laufer, Solon Barocas, Manish Raghavan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add data-adaptive CEI bounds (extra assumptions) to the stopping bound | Anytime-valid LDA search stopping bound · Adult/Folktables/HMDA | upper bound tightness | No clear effect |
| add extra distributional assumptions to the bound | Anytime-valid LDA search stopping bound · Adult/Folktables/HMDA | upper bound tightness | No clear effect |
| add Fairlearn fairness regularization to the models | Retrained models · Adult/Folktables/HMDA | upper bound on marginal gain | Improved |
| lower Fairlearn target disparate impact to 0.05 (ablation) | Fairlearn methods · Adult/Folktables/HMDA | marginal gain from retraining: 0.05 | Improved |
| use finite-data version instead of infinite-data version | Anytime-valid upper bound · Miscoverage | miscoverage rate: less than 10 events -> significantly above target 0.05 rate | Hurt |
Closest papers by meaning
- Audits Under Resource, Data, and Access Constraints: Scaling Laws For Less Discriminatory Alternatives (NeurIPS 2025 · similarity 0.85)
- Learning Fair Policies for Multi-Stage Selection Problems from Observational Data (AAAI 2024 · similarity 0.85)
- Reconciling Predictive and Statistical Parity: A Causal Approach (AAAI 2024 · similarity 0.84)
- Size-adaptive Hypothesis Testing for Fairness (NeurIPS 2025 · similarity 0.84)
- Mind the Gap: A Causal Perspective on Bias Amplification in Prediction & Decision-Making (NeurIPS 2024 · similarity 0.84)
- Fairness-Accuracy Trade-Offs: A Causal Perspective (AAAI 2025 · similarity 0.83)
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.
