Research map / Learning theory and algorithms / Kernels and regression theory
Kernel Feature Selection via Conditional Covariance Minimization
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 |
|---|---|---|---|
| Use conditional covariance minimization (CCM) instead of the compared feature-selection algorithms | Feature selection with downstream kernel SVM · 12 real-world benchmark tasks | classification accuracy | Helped |
| Use CCM instead of BAHSIC | Feature selection with downstream kernel SVM · 12 real-world benchmark tasks | classification accuracy | Helped |
| Use CCM instead of the compared feature-selection algorithms | Synthetic binary classification · Synthetic binary classification | median rank of true features | Helped |
| Use CCM instead of the compared feature-selection algorithms | Synthetic 4-way classification · 3-dimensional XOR | median rank of true features | Helped |
| Use CCM instead of BAHSIC | Feature selection · Synthetic 3-dimensional XOR | median rank of true features | Helped |
| Use CCM instead of the compared feature-selection algorithms | Synthetic feature selection · Additive nonlinear regression | median rank of true features | No clear effect |
| Replace the initial relaxed objective with the soft feature-count constraint objective (ablation) | CCM · Synthetic data | median rank of true features | No clear effect |
| Replace the initial relaxed objective with the soft matrix-inverse constraint objective (ablation) | CCM · Synthetic data | median rank of true features | No clear effect |
| Use the soft feature-count constraint rather than the soft matrix-inverse constraint (ablation) | CCM · Synthetic data | robustness to hyperparameter changes | Helped |
Closest papers by meaning
- Few-shot Learning for Feature Selection with Hilbert-Schmidt Independence Criterion (NeurIPS 2022 · similarity 0.87)
- Practical Kernel Selection for Kernel-based Conditional Independence Test (NeurIPS 2025 · similarity 0.87)
- Kernel Functional Optimisation (NeurIPS 2021 · similarity 0.87)
- Estimating Conditional Mutual Information for Dynamic Feature Selection (ICLR 2024 · similarity 0.86)
- SAND: One-Shot Feature Selection with Additive Noise Distortion (ICML 2025 · similarity 0.85)
- Contextual Feature Selection with Conditional Stochastic Gates (ICML 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.