Research map / Learning theory and algorithms / Fairness
Fair Regression: Quantitative Definitions and Reduction-Based Algorithms
Alekh Agarwal, Miroslav Dudik, Zhiwei Steven Wu
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 |
|---|---|---|---|
| Apply the proposed statistical-parity fair regression method instead of unconstrained regression | Regression on Adult, Law School, and Communities & Crime · Adult, Law School, Communities & Crime | statistical-parity disparity | Improved |
| Apply the proposed fair regression method instead of SEO | Least-squares regression · Law School and Communities & Crime; adult datasets where applicable | relative test loss and disparity tradeoff | Improved |
| Use the proposed fair regression method instead of fair classification | Logistic regression · Adult and adult subsampled | relative test loss | Mixed |
| Use the proposed method rather than fair classification | Logistic regression · Adult | relative test loss | Hurt |
| Use the proposed method rather than fair classification | Logistic regression · Adult subsampled | relative test loss | No clear effect |
| Use least-squares or logistic-loss reduction instead of cost-sensitive classification reduction | Statistical-parity fair regression · Evaluated datasets | fairness-accuracy tradeoff | No clear effect |
| Use least-squares or logistic-loss reduction instead of cost-sensitive classification reduction | Statistical-parity fair regression · Evaluated datasets | runtime | Improved |
| Reduce to a matching loss rather than a different loss | Fair regression reduction oracles · Adult and Adult subsampled | relative test loss | Improved |
| Use the LS/LR reduction heuristics rather than the cost-sensitive reduction | Statistical-parity fair regression · Adult | relative test loss | Hurt |
| Enforce statistical parity across all score thresholds rather than only at the classification threshold | Proposed fair regression versus fair classification · Logistic-regression datasets | statistical-parity disparity | Improved |
| Use the proposed method on Communities & Crime | Statistical-parity fair regression · Communities & Crime | relative test loss | Mixed |
Closest papers by meaning
- Reconciling Predictive and Statistical Parity: A Causal Approach (AAAI 2024 · similarity 0.90)
- Geometry of Relaxed Fair Regression: A Unified Framework for Aware and Unaware Settings (NeurIPS 2026 · similarity 0.90)
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.89)
- Demographic parity in regression and classification within the unawareness framework (NeurIPS 2026 · similarity 0.89)
- The Fragility of Fairness: Causal Sensitivity Analysis for Fair Machine Learning (NeurIPS 2024 · similarity 0.88)
- FairPFN: A Tabular Foundation Model for Causal Fairness (ICML 2025 · 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.
