Research map / Learning theory and algorithms / Fairness
Convex Formulations for Fair Principal Component Analysis
Matt Olfat, Anil Aswani
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 mean and covariance constraints to PCA | FPCA · UC Irvine datasets (15 datasets) | linear SVM Δ-fairness: PCA -> FPCA - Both Con.: Adult Income 0.54 -> 0.07; Wine Quality 0.97 -> 0.06 | Improved |
| Use mean constraint only rather than unconstrained PCA (ablation) | FPCA · UC Irvine datasets | linear SVM Δ-fairness: Adult Income 0.54 -> 0.14; Wine Quality 0.97 -> 0.21 | Improved |
| Add covariance constraints to mean-constrained FPCA (ablation) | FPCA - Mean Con. · UC Irvine datasets | linear SVM Δ-fairness: Adult Income 0.14 -> 0.07; Magic 0.11 -> 0.12 | Mixed |
| Add both fairness constraints rather than unconstrained PCA (ablation) | PCA · NHANES activity clustering by age | standard deviation of cluster age proportions: 14.87% -> 1.79% | Improved |
| Add mean constraint rather than unconstrained PCA (ablation) | PCA · NHANES activity clustering by age | standard deviation of cluster age proportions: 14.87% -> 2.46% | Improved |
| Tighten covariance bound μ (ablation) | FPCA with both constraints · Wine Quality | linear SVM Δ-fairness | Improved |
| Increase μ beyond 0.1 (ablation) | FPCA with both constraints · Wine Quality | linear SVM Δ-fairness: 0.1 | No clear effect |
| Add covariance constraints to mean-only FPCA (ablation) | FPCA · Synthetic data | protected-group separability | Improved |
Closest papers by meaning
- Multi-Criteria Dimensionality Reduction with Applications to Fairness (NeurIPS 2019 · similarity 0.91)
- The Price of Fair PCA: One Extra dimension (NeurIPS 2018 · similarity 0.90)
- Fast and Efficient MMD-Based Fair PCA via Optimization over Stiefel Manifold (AAAI 2022 · similarity 0.90)
- Understanding Fairness and Prediction Error through Subspace Decomposition and Influence Analysis (NeurIPS 2025 · similarity 0.86)
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.85)
- Kernel-based Equalized Odds: A Quantification of Accuracy-Fairness Trade-off in Fair Representation Learning (NeurIPS 2025 · similarity 0.84)
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.
