Research map / Learning theory and algorithms / Kernels and regression theory
Convergence guarantees for kernel-based quadrature rules in misspecified settings
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 kernel quadrature designed for smoothness r when the integrand has lesser smoothness s | Kernel quadrature with random points · General RKHS misspecified setting | integration error convergence rate: O(n^{-θb}) | Helped |
| Use randomized lattice rules or Bach quadrature despite lesser integrand smoothness | Randomized lattice rules and Bach quadrature · Korobov spaces | integration error convergence rate: O(n^{-αθ+ξ}) | Helped |
| Construct deterministic quadrature points assuming smoothness r for an integrand of lesser smoothness s | Deterministic kernel quadrature in Sobolev spaces · Sobolev W₂ˢ, s ≤ r | integration error convergence rate: O(n^{-bs/r}) | Helped |
| Use points with the optimal rate in W₂ʳ without knowing the integrand smoothness s | Deterministic Sobolev quadrature · Sobolev W₂ˢ | integration error convergence rate: O(n^{-s/d}) | Helped |
| Use deterministic points satisfying the separation-radius condition | Sobolev kernel quadrature · Sobolev W₂ˢ, s ≤ r | integration error convergence rate: O(n^{-bs/r}) | Helped |
| Apply unrandomized fast-CBC QMC lattice rules to integrands less smooth than assumed | QMC lattice rules · Korobov space, d = 2; s = 1, 2, 3 | integration error convergence rate | Mixed |
Closest papers by meaning
- Kernel quadrature with DPPs (NeurIPS 2019 · similarity 0.90)
- An analysis of Ermakov-Zolotukhin quadrature using kernels (NeurIPS 2021 · similarity 0.89)
- On the Sampling Problem for Kernel Quadrature (ICML 2017 · similarity 0.87)
- Positively Weighted Kernel Quadrature via Subsampling (NeurIPS 2022 · similarity 0.87)
- Quadrature-based features for kernel approximation (NeurIPS 2018 · similarity 0.86)
- Kernel conditional tests from learning-theoretic bounds (NeurIPS 2025 · similarity 0.84)
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.