Research map / Learning theory and algorithms / Kernels and regression theory
Random Fourier Features via Fast Surrogate Leverage Weighted Sampling
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 surrogate leverage weighted sampling instead of leverage-RFF | RFF for kernel ridge regression · EEG | classification accuracy (%): 90.01±0.27 -> 91.02±0.32 | Helped |
| Use surrogate leverage weighted sampling instead of leverage-RFF | RFF for kernel ridge regression · cod-RNA | classification accuracy (%): 93.66±0.03 -> 93.48±0.04 | No clear effect |
| Use surrogate leverage weighted sampling instead of leverage-RFF | RFF for kernel ridge regression · covtype | classification accuracy (%): 82.90±0.12 -> 82.18±0.10 | Hurt |
| Use surrogate leverage weighted sampling instead of leverage-RFF | RFF for kernel ridge regression · magic04 | classification accuracy (%): 82.59±0.29 -> 82.55±0.55 | No clear effect |
| Replace leverage-RFF sampling with surrogate leverage weighted sampling | RFF feature generation · EEG | time cost for generating random features (sec.): 3.91±0.09 -> 0.96±0.03 | Helped |
| Replace leverage-RFF sampling with surrogate leverage weighted sampling | RFF feature generation · cod-RNA | time cost for generating random features (sec.): 13.3±0.23 -> 7.78±0.09 | Helped |
| Replace leverage-RFF sampling with surrogate leverage weighted sampling | RFF feature generation · covtype | time cost for generating random features (sec.): 46.5±2.20 -> 28.6±1.58 | Helped |
| Replace leverage-RFF sampling with surrogate leverage weighted sampling | RFF feature generation · magic04 | time cost for generating random features (sec.): 2.61±0.15 -> 0.87±0.02 | Helped |
| Use data-dependent leverage-weighted sampling rather than RFF or QMC | Random Fourier features · EEG | classification accuracy (%): At 128d, Ours 91.02±0.32; RFF 79.79±0.49; QMC 79.51±0.47 | Helped |
| Use data-dependent leverage-weighted sampling rather than RFF or QMC | Random Fourier features · cod-RNA | classification accuracy (%): At 64d, Ours 93.99±0.06; RFF 87.30±0.03; QMC 87.32±0.00 | Helped |
| Use data-dependent leverage-weighted sampling rather than RFF or QMC | Random Fourier features · covtype | classification accuracy (%): At 16d: Ours 82.18±0.10; RFF 82.42±0.10; QMC 82.37±0.07 | No clear effect |
| Use data-dependent leverage-weighted sampling rather than RFF or QMC | Random Fourier features · magic04 | classification accuracy (%): At 128d, Ours 82.55±0.55; RFF 81.10±0.37; QMC 81.08±0.29 | Helped |
Closest papers by meaning
- Learning Adaptive Random Features (AAAI 2019 · similarity 0.90)
- Random Fourier Features for Kernel Ridge Regression: Approximation Bounds and Statistical Guarantees (ICML 2017 · similarity 0.90)
- Fourier Sparse Leverage Scores and Approximate Kernel Learning (NeurIPS 2020 · similarity 0.89)
- TRF: Learning Kernels with Tuned Random Features (AAAI 2022 · similarity 0.89)
- Recursive Sampling for the Nystrom Method (NeurIPS 2017 · similarity 0.88)
- Kernel Regression in Structured Non-IID Settings: Theory and Implications for Denoising Score Learning (NeurIPS 2025 · 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.