Research map / Learning theory and algorithms / Fairness
Extending Fair Null-Space Projections for Continuous Attributes to Kernel Methods
Felix Störck, Fabian Hinder, Barbara Hammer
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 |
|---|---|---|---|
| apply kernel null-space projection (FKD) to SVR instead of other fair methods | SVR-FKD · ACSIncome (Montana, age protected) | MAE vs GDP/HGR/PF Pareto front | Improved |
| apply kernel null-space projection to KRR compared with KRR-FKL | KRR-FKD · ACSIncome (Montana) | MAE vs fairness Pareto front | Hurt |
| use SVR-FKD instead of NN-HGR and KRR-FKL under strong regularization | SVR-FKD · Communities & Crimes (HGR measure) | MAE vs HGR | Hurt |
| use SVR-FKD under weak regularization | SVR-FKD · Communities & Crimes (GDP) | MAE vs GDP | Improved |
| apply SVR-FKD where other methods match the dummy regressor | SVR-FKD · ACSTravelTime (Montana) | MAE vs fairness | Improved |
| protect both white and black population percentage instead of only black (ablation) | SVR-FKD · Communities & Crimes | GDP fairness vs MAE | Improved |
| increase ridge regularization alpha in the projection (ablation) | KRR-FKD · Communities & Crimes | MAE | Hurt |
| use kernel-space projection instead of linear removal on input data (ablation) | SVR-FKD vs SVR-INPL · ACSIncome | fairness at matched iterations m | Improved |
| use Nystroem approximation of the kernel inverse (25% to 75% of components) (ablation) | SVR-FKD · Communities & Crimes | MAE vs GDP | No clear effect |
Closest papers by meaning
- Fairness without Demographics through Shared Latent Space-Based Debiasing (AAAI 2024 · similarity 0.87)
- A Sequentially Fair Mechanism for Multiple Sensitive Attributes (AAAI 2024 · similarity 0.87)
- Kernel-based Equalized Odds: A Quantification of Accuracy-Fairness Trade-off in Fair Representation Learning (NeurIPS 2025 · similarity 0.86)
- Understanding Fairness and Prediction Error through Subspace Decomposition and Influence Analysis (NeurIPS 2025 · similarity 0.86)
- The Fragility of Fairness: Causal Sensitivity Analysis for Fair Machine Learning (NeurIPS 2024 · similarity 0.86)
- 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.
