Research map / Learning theory and algorithms / Kernels and regression theory
Efficient Numerical Integration in Reproducing Kernel Hilbert Spaces via Leverage Scores Sampling
Antoine Chatalic, Nicolas Schreuder, Ernesto De Vito, Lorenzo Rosasco
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 |
|---|---|---|---|
| Use approximate ridge leverage score sampling instead of uniform sampling of nodes (ablation) | Nystrom-based RKHS quadrature · OpenML datasets, Gaussian kernel | worst-case error vs number of nodes m | Improved |
| Use approximate ridge leverage score sampling instead of uniform sampling of nodes (ablation) | Nystrom-based RKHS quadrature · Periodic Sobolev spaces (d=1,s=1 and d=2,s=3) | worst-case error | No clear effect |
| Use approximate ridge leverage score sampling instead of uniform sampling of nodes (ablation) | Nystrom-based RKHS quadrature · OpenML datasets, Laplacian kernel | worst-case error | No clear effect |
| Use proposed quadrature instead of greedy and DPP baselines at equal runtime | Nystrom-based RKHS quadrature · Periodic Sobolev spaces | error vs runtime tradeoff | Improved |
| Use proposed quadrature instead of greedy and DPP baselines at fixed node count m | Nystrom-based RKHS quadrature · Periodic Sobolev spaces | error at fixed m | Hurt |
| Use proposed quadrature instead of other randomized and greedy baselines at equal runtime | Nystrom-based RKHS quadrature · OpenML datasets, Gaussian kernel | error vs runtime tradeoff | Improved |
| Use greedy det(K_m) maximization (space-filling) instead of data-adaptive selection | Greedy baseline · OpenML datasets, Gaussian kernel | accuracy | Hurt |
Closest papers by meaning
- Recursive Sampling for the Nystrom Method (NeurIPS 2017 · similarity 0.83)
- Kernelized Cumulants: Beyond Kernel Mean Embeddings (NeurIPS 2023 · similarity 0.82)
- Relating Leverage Scores and Density using Regularized Christoffel Functions (NeurIPS 2018 · similarity 0.82)
- Sampling-based Nyström Approximation and Kernel Quadrature (ICML 2023 · similarity 0.82)
- A Data-Adaptive RKHS Prior for Bayesian Learning of Kernels in Operators (JMLR 2024 · similarity 0.81)
- Guided Diffusion Sampling on Function Spaces with Applications to PDEs (NeurIPS 2025 · similarity 0.80)
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.
