Research map / Learning theory and algorithms / Fairness
A Reductions Approach to Fair Classification
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 exponentiated-gradient reduction instead of the evaluated baseline methods | Binary fair classification · Adult, COMPAS, Law School, Dutch census; demographic parity and equalized odds test sets | classification error at comparable fairness constraint violation | Helped |
| Use score-based post-processing instead of the exponentiated-gradient reduction | Binary fair classification · Adult, COMPAS, Law School, Dutch census; demographic parity and equalized odds test sets | classification error at statistically indistinguishable disparity | Hurt |
| Use reweighting instead of the exponentiated-gradient reduction | Demographic parity classification · Adult, COMPAS, Law School, Dutch census test sets | fairness constraint violation | Hurt |
| Use relabeling instead of the exponentiated-gradient reduction | Demographic parity classification with binary protected attributes · Adult, COMPAS, Law School, Dutch census test sets where applicable | accuracy-fairness Pareto optimality | Hurt |
| Use grid-search reduction instead of exponentiated-gradient reduction | Binary fair classification · Adult, COMPAS, Law School, Dutch census test sets | classification error and fairness constraint violation | No clear effect |
| Apply fairness constraints on the Dutch census data | Fair classification · Dutch census | classification error | Hurt |
Closest papers by meaning
- Leveraging Labeled and Unlabeled Data for Consistent Fair Binary Classification (NeurIPS 2019 · similarity 0.92)
- Demystifying the Optimal Fair Classifier in Multi-Class Classification (ICML 2026 · similarity 0.90)
- Post-hoc bias scoring is optimal for fair classification (ICLR 2024 · similarity 0.89)
- Efficient Fairness-Performance Pareto Front Computation (NeurIPS 2025 · similarity 0.88)
- Individual Fairness In Strategic Classification (NeurIPS 2025 · similarity 0.88)
- Demographic parity in regression and classification within the unawareness framework (NeurIPS 2026 · 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.