Research map / Learning theory and algorithms / Kernels and regression theory
On Data-Dependent Random Features for Improved Generalization in Supervised Learning
Shahin Shahrampour, Ahmad Beirami, Vahid Tarokh
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 |
|---|---|---|---|
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Random Fourier features with a Gaussian kernel · Buzz prediction on Twitter | test error (%): 6.09 (7e-2) -> 4.65 (4e-2) | Improved |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Random Fourier features with a Gaussian kernel · Online news popularity | test error (%): 3.08 (5e-2) -> 1.63 (4e-2) | Improved |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Random Fourier features with a Gaussian kernel · Adult | test error (%): 17.37 (6e-2) -> 16.16 (2e-2) | Improved |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Random Fourier features with a Gaussian kernel · MNIST | test error (%): 7.53 (1.8e-1) -> 7.28 (3e-2) | Improved |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Data-dependent random features · Buzz prediction on Twitter | test error (%): 5.23 (6e-2) -> 4.65 (4e-2) | Improved |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Data-dependent random features · Online news popularity | test error (%): 2.07 (5e-2) -> 1.63 (4e-2) | Improved |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Data-dependent random features · Adult | test error (%): 16.34 (2e-2) -> 16.16 (2e-2) | No clear effect |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Data-dependent random features · MNIST | test error (%): 7.59 (1.7e-1) -> 7.28 (3e-2) | Improved |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Data-dependent random features · Year prediction | test error (%): 7.40 (1.4e-2) -> 7.15 (1.9e-2) | Improved |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Data-dependent random features · Letter recognition | test error (%): 7.17 (8e-2) -> 6.83 (7e-2) | Improved |
| Select features by the empirical score |Ŝ(ω)| from an initial pool | Data-dependent random features · Hand movement | test error (%): 7.24 (1.2e-2) -> 6.14 (2.7e-3) | Improved |
Closest papers by meaning
- Learning Kernels with Random Features (NeurIPS 2016 · similarity 0.89)
- Generalization Properties of Learning with Random Features (NeurIPS 2017 · similarity 0.88)
- Random Features for Kernel Approximation: A Survey on Algorithms, Theory, and Beyond (TPAMI 2021 · similarity 0.86)
- An Empirical Study on The Properties of Random Bases for Kernel Methods (NeurIPS 2017 · similarity 0.85)
- Characterizing the Generalization Error of Random Feature Regression with Arbitrary Data-Augmentation (NeurIPS 2026 · similarity 0.84)
- General Graph Random Features (ICLR 2024 · 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.
