Research map / Learning theory and algorithms / Kernels and regression theory
Error Analysis of Generalized Nyström Kernel Regression
Hong Chen, Haifeng Xia, Heng Huang, Weidong Cai
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 |
|---|---|---|---|
| Epanechnikov instead of Gaussian kernel (ablation) | GNKR · f3(x)=sign(x), size 1000 | RMSE: 0.28742 -> 0.14726 | Improved |
| Epanechnikov instead of Gaussian kernel (ablation) | GNKR · f4, size 1000 | RMSE: 0.35139 -> 0.18560 | Improved |
| Epanechnikov instead of Gaussian kernel (ablation) | GNKR · f1 smooth, size 1000 | RMSE: 0.02908 -> 0.08024 | Hurt |
| Epanechnikov instead of Gaussian kernel (ablation) | GNKR · f2 smooth, size 1000 | RMSE: 0.03395 -> 0.03440 | Hurt |
| Epanechnikov instead of Gaussian kernel (ablation) | GNKR · CASP, 100 samples | RMSE (x10^-3): 9.238 -> 4.196 | Improved |
| Epanechnikov instead of Gaussian kernel (ablation) | GNKR · Year Prediction, 100 samples | RMSE (x10^-3): 8.802 -> 6.842 | Improved |
| Epanechnikov instead of Gaussian kernel (ablation) | GNKR · census-house, 100 samples | RMSE (x10^-3): 111.083 -> 99.535 | Improved |
| column-norm subsampling instead of uniform (ablation) | GNKR with Epanechnikov · f2, size 300 | RMSE: 0.04786 -> 0.04607 | Improved |
| column-norm subsampling instead of uniform (ablation) | GNKR with Epanechnikov · f3, size 300 | RMSE: 0.16170 -> 0.16500 | Hurt |
| column-norm subsampling instead of uniform (ablation) | GNKR with Gaussian · CASP, 100 samples | RMSE (x10^-3): 9.238 -> 9.196 | Improved |
| column-norm subsampling instead of uniform (ablation) | GNKR with Epanechnikov · CASP, 100 samples | RMSE (x10^-3): 4.196 -> 4.249 | Hurt |
| increase sample size from 300 to 1000 | E-GNKR-US · f1 | RMSE: 0.10159 -> 0.08024 | Improved |
Closest papers by meaning
- Generalization Properties of hyper-RKHS and its Applications (JMLR 2021 · similarity 0.90)
- Characterizing Overfitting in Kernel Ridgeless Regression Through the Eigenspectrum (ICML 2024 · similarity 0.88)
- A Comprehensive Analysis on the Learning Curve in Kernel Ridge Regression (NeurIPS 2024 · similarity 0.87)
- Kernel Regression in Structured Non-IID Settings: Theory and Implications for Denoising Score Learning (NeurIPS 2025 · similarity 0.87)
- Generalization error of spectral algorithms (ICLR 2024 · similarity 0.87)
- Overfitting Behaviour of Gaussian Kernel Ridgeless Regression: Varying Bandwidth or Dimensionality (NeurIPS 2024 · 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.
