Research map / Learning theory and algorithms / Fairness
Fair Streaming Principal Component Analysis: Statistical and Algorithmic Viewpoint
Junghyun Lee, Hanseul Cho, Se-Young Yun, Chulhee Yun
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 |
|---|---|---|---|
| streaming FNPM instead of offline fair PCA with the same mean-constraint | FNPM · Adult Income k=2 | %Var: 5.74 -> 5.74 | No clear effect |
| streaming FNPM instead of offline fair PCA with m=15 | FNPM · Adult Income k=2 | %Var: 4.04 -> 4.07 | No clear effect |
| streaming FNPM instead of offline fair PCA with m=50 | FNPM · Adult Income k=10 | %Var: 11.34 -> 11.34 | No clear effect |
| increase constraint count from mean to m=15 then m=50 | FNPM · Adult Income k=2 | %Var: 5.74 -> 4.07 -> 2.64 | Hurt |
| tighten fairness by using m=50 instead of mean constraint | FNPM · Adult Income k=10 | kernel SVM Delta_DP: 0.18 -> 0.11 | Improved |
| tighten fairness by using m=50 instead of mean constraint | FNPM · Adult Income k=10 | kernel SVM accuracy %: 88.53 -> 82.85 | Hurt |
| apply FNPM mean constraint instead of plain PCA | FNPM · Adult Income k=2 | MMD^2: 0.374 -> 0.002 | Improved |
| apply FNPM mean constraint instead of plain PCA | FNPM · Adult Income k=2 | kernel SVM Delta_DP: 0.19 -> 0.07 | Improved |
| apply FNPM mean constraint instead of plain PCA | FNPM · COMPAS k=2 | MMD^2: 0.057 -> 0.008 | Improved |
| apply FNPM mean constraint instead of plain PCA | FNPM · German Credit k=2 | %Var: 11.13 -> 10.39 | No clear effect |
| FNPM mean constraint versus Ravfogel et al. 2020 baseline | FNPM · Adult Income k=2 | %Var: 1.91 -> 5.74 | Improved |
| FNPM mean constraint versus Samadi et al. 2018 baseline | FNPM · Adult Income k=2 | kernel SVM Delta_DP: 0.15 -> 0.07 | Improved |
Closest papers by meaning
- Fast and Efficient MMD-Based Fair PCA via Optimization over Stiefel Manifold (AAAI 2022 · similarity 0.89)
- Low Precision Streaming PCA (NeurIPS 2025 · similarity 0.85)
- Oja's Algorithm for Streaming Sparse PCA (NeurIPS 2024 · similarity 0.85)
- Capturing the denoising effect of PCA via compression ratio (NeurIPS 2024 · similarity 0.84)
- Efficient Sparse PCA via Block-Diagonalization (ICLR 2025 · similarity 0.83)
- Rethinking PCA Through Duality (NeurIPS 2025 · similarity 0.83)
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.
