Research map / Learning theory and algorithms / Fairness
Fairness guarantees in multi-class classification with demographic parity
Christophe Denis, Romuald Elie, Mohamed Hebiri, François Hu
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 exact fairness calibration (epsilon = 0) to the random forest classifier | Random forest on synthetic data · Synthetic Gaussian mixture | unfairness | Improved |
| add the fairness constraint (decrease epsilon) | epsilon-fair classifier · Synthetic Gaussian mixture | accuracy | Hurt |
| increase epsilon to 0.2 (ablation) | epsilon-fair classifier · Synthetic Gaussian mixture | unfairness and accuracy | No clear effect |
| split data so frequencies are estimated on independent unlabeled data (ablation) | epsilon-fairness calibration · Synthetic, p=0.75 | unfairness and accuracy | Improved |
| use epsilon-fair calibration instead of fair-adversarial | binary classification · DRUG and CRIME (K=2) | accuracy and unfairness | Improved |
| use epsilon-fair calibration instead of fair-projection | multi-class classification · DRUG and CRIME (multi-class) | unfairness | Improved |
| use epsilon-fair calibration instead of fair-transport | multi-class classification · DRUG and CRIME (multi-class) | accuracy-fairness trade-off | No clear effect |
Closest papers by meaning
- Demystifying the Optimal Fair Classifier in Multi-Class Classification (ICML 2026 · similarity 0.92)
- Demographic parity in regression and classification within the unawareness framework (NeurIPS 2026 · similarity 0.89)
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.89)
- On the Maximal Local Disparity of Fairness-Aware Classifiers (ICML 2024 · similarity 0.88)
- Geometry of Relaxed Fair Regression: A Unified Framework for Aware and Unaware Settings (NeurIPS 2026 · similarity 0.88)
- Efficient Fairness-Performance Pareto Front Computation (NeurIPS 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.
