Research map / Learning theory and algorithms / Fairness
Fair Bayes-Optimal Classifiers Under Predictive Parity
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 |
|---|---|---|---|
| apply FairBayes-DPP adaptive thresholding instead of unconstrained classifier | logistic regression classifier · Synthetic, P(Y=1|A=1)=0.6 | DPP: 0.113 -> 0.038 | Helped |
| apply FairBayes-DPP | logistic regression classifier · Synthetic, P(Y=1|A=1)=0.5 | DPP: 0.081 -> 0.042 | Helped |
| apply FairBayes-DPP | logistic regression classifier · Synthetic, P(Y=1|A=1)=0.2 | DPP: 0.046 -> 0.049 | No clear effect |
| apply FairBayes-DPP | logistic regression classifier · Synthetic, P(Y=1|A=1)=0.6 | accuracy: 0.781 -> 0.778 | No clear effect |
| impose predictive parity on true Bayes-optimal classifier | Bayes-optimal classifier · Synthetic, P(Y=1|A=1)=0.5 | accuracy: 0.777 -> 0.775 | No clear effect |
| apply FairBayes-DPP | Adult and COMPAS neural network · Adult | DPP | Helped |
| apply FairBayes-DPP | Adult and COMPAS neural network · COMPAS | DPP | Helped |
| apply FairBayes-DPP | Adult and COMPAS neural network · Adult and COMPAS | accuracy | No clear effect |
| apply FairBayes-DPP | ResNet50 · CelebA Bangs | DPP: 0.069 -> 0.007 | Helped |
| apply FairBayes-DPP | ResNet50 · CelebA Attractive | DPP: 0.169 -> 0.075 | Helped |
| apply FairBayes-DPP | ResNet50 · CelebA Big Nose | DPP: 0.145 -> 0.083 | Helped |
| apply FairBayes-DPP | ResNet50 · CelebA Big Lips | accuracy: 0.717 -> 0.706 | Hurt |
Closest papers by meaning
- Reconciling Predictive and Statistical Parity: A Causal Approach (AAAI 2024 · similarity 0.91)
- The Fragility of Fairness: Causal Sensitivity Analysis for Fair Machine Learning (NeurIPS 2024 · similarity 0.88)
- Arbitrariness and Social Prediction: The Confounding Role of Variance in Fair Classification (AAAI 2024 · similarity 0.88)
- Demographic parity in regression and classification within the unawareness framework (NeurIPS 2026 · similarity 0.88)
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.88)
- Post-hoc bias scoring is optimal for fair classification (ICLR 2024 · 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.