Research map / Learning theory and algorithms / Fairness
Fair Densities via Boosting the Sufficient Statistics of Exponential Families
Alexander Soen, Hisham Husain, Richard Nock
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 |
|---|---|---|---|
| exact leverage (sr0=1) vs relative leverage (sr0=0.9) in FBDE (ablation) | FBDE · COMPAS and Adult | SR vs KL trade-off: COMPAS SR .988 vs .896; KL .135 vs .127 | Improved |
| FBDE boosting (preprocessing) versus Celis max-entropy baseline | FBDE · Adult (S=sex) | KL: 0.122 vs 0.182 | Improved |
| Celis max-entropy baseline vs FBDE | Celis · COMPAS and Adult | SR: 0.992 / 0.979 vs ~0.96-0.99 | Improved |
| TabFairGAN pre-processing | TabFairGAN · COMPAS and Adult | SR and RR: COMPAS SR .727 with std .100; Adult KL 1.68 | Hurt |
| FBDE with relative leverage vs exact on continuous Minneapolis data | FBDE · Minneapolis (continuous) | RR: RR 0.847 vs theoretical sqrt(tau) approx 0.894 | No clear effect |
| Decision tree WL based FBDE, data fairness (SR) on COMPAS | FBDE · COMPAS and Adult | prediction fairness SR_clf: COMPAS SR_clf .959 for FBDE M_a; Adult .818 for FBDE M_a vs .919 for Celis | Mixed |
Closest papers by meaning
- fairret: a Framework for Differentiable Fairness Regularization Terms (ICLR 2024 · similarity 0.85)
- Differentially Private Post-Processing for Fair Regression (ICML 2024 · similarity 0.85)
- A Comprehensive View of Fairness through Distributional Stability (NeurIPS 2026 · similarity 0.85)
- Fairness-aware Bayes Optimal Functional Classification (NeurIPS 2025 · similarity 0.85)
- Fairness Overfitting in Machine Learning: An Information-Theoretic Perspective (ICML 2025 · similarity 0.85)
- Fairness without Harm: An Influence-Guided Active Sampling Approach (NeurIPS 2024 · similarity 0.85)
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.
