Research map / Learning theory and algorithms / Kernels and regression theory
Quadrature-based features for kernel approximation
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 degree (3,3) spherical-radial quadrature features instead of standard random-feature sampling | Kernel approximation · Most tested kernels and datasets | relative Frobenius kernel approximation error | Helped |
| Use degree (3,3) spherical-radial quadrature features instead of standard random-feature sampling | SVMs with approximate kernels · Classification/regression datasets | accuracy or R² | Mixed |
| Use butterfly structured orthogonal matrices for the quadrature mapping | Quadrature-based features · Feature mapping across increasing input dimensions | mapping time | Helped |
| Use independently sampled radial values for each simplex vertex rather than one shared radial value (ablation) | Degree (3,3) spherical-radial rule · Kernel approximation | approximation quality | No clear effect |
| Use butterfly matrices instead of QR-generated random orthogonal matrices | Quadrature-based features · Random orthogonal matrix generation | performance | No clear effect |
| Use QMC rather than RFF sampling | Kernel approximation · Most tested settings | relative Frobenius kernel approximation error | No clear effect |
| Use Gaussian quadrature rather than RFF sampling | Kernel approximation · Reported comparable-dimension settings | relative Frobenius kernel approximation error | No clear effect |
Closest papers by meaning
- Gaussian Quadrature for Kernel Features (NeurIPS 2017 · similarity 0.90)
- Learning Kernels with Random Features (NeurIPS 2016 · similarity 0.90)
- Random Features for Shift-Invariant Kernels with Moment Matching (AAAI 2017 · similarity 0.89)
- Quasi-Monte Carlo Features for Kernel Approximation (ICML 2024 · similarity 0.88)
- Random features for Grassmannian kernel approximation with bounded rank-one projections (NeurIPS 2026 · similarity 0.88)
- Manifold Random Features (NeurIPS 2026 · similarity 0.87)
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.