Research map / Learning theory and algorithms / Kernels and regression theory
On Fast Leverage Score Sampling and Optimal Learning
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 BLESS instead of SQUEAK for approximate leverage-score sampling | Leverage-score sampling · SUSY (n = 70,000, λ = 10⁻⁵) | Time: 52 -> 17 | Helped |
| Use BLESS instead of SQUEAK for approximate leverage-score sampling | Leverage-score sampling · SUSY (n = 70,000, λ = 10⁻⁵) | R-ACC: 1.06 -> 1.06 | No clear effect |
| Sample without replacement (ablation) | BLESS · SUSY (n = 70,000, λ = 10⁻⁵) | R-ACC: 1.06 -> 1.06 | No clear effect |
| Sample without replacement (ablation) | BLESS · SUSY (n = 70,000, λ = 10⁻⁵) | 5th / 95th quantile of R-ACC: 0.57 / 2.03 -> 0.73 / 1.50 | Helped |
| Use BLESS instead of uniform sampling | Approximate leverage-score sampling · SUSY (n = 70,000, λ = 10⁻⁵) | R-ACC: 1.09 -> 1.06 | Helped |
| Use leverage-score sampling rather than uniform sampling | FALKON · SUSY | Time to reach the same accuracy: 610 seconds -> 160 seconds | Helped |
| Use leverage-score sampling rather than uniform sampling | FALKON · HIGGS | Time to reach better accuracy: 2.7 hours -> 1.4 hours | Helped |
Closest papers by meaning
- Generalized Leverage Scores: Geometric Interpretation and Applications (ICML 2022 · similarity 0.90)
- Improved Active Learning via Dependent Leverage Score Sampling (ICLR 2024 · similarity 0.86)
- Turnstile $\ell_p$ leverage score sampling with applications (ICML 2024 · similarity 0.85)
- Symmetric Matrix Completion with ReLU Sampling (ICML 2024 · similarity 0.84)
- Provably Accurate Shapley Value Estimation via Leverage Score Sampling (ICLR 2025 · similarity 0.83)
- Efficient Leverage Score Sampling for Tensor Train Decomposition (NeurIPS 2024 · similarity 0.83)
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.