Research map / Learning theory and algorithms / Kernels and regression theory
Ridge Regression: Structure, Cross-Validation, and Sketching
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 primal orthogonal sketching to reduce the sample size to half | Ridge regression · Primal orthogonal sketching simulation | MSE relative to ridge regression: 1.05 | Hurt |
| Scale the K-fold CV-selected λ by (K-1)/K | Ridge regression · K-fold CV simulation (γ=0.7) | test error: 1.513 -> 1.510 | Helped |
| Scale the K-fold CV-selected λ by (K-1)/K | Ridge regression · K-fold CV simulation (γ=5) | test error: 8.92 to 8.89 | Helped |
| Use the bias-corrected CV-selected λ | Ridge regression · Million Song Dataset | test error: 0.00024 | Helped |
| Use the bias-corrected CV-selected λ | Ridge regression · Flights dataset | test error: 0.0022 | Helped |
| Use marginal regression instead of optimally tuned ridge | Regression with α²=σ²=1 and γ=0.7 · Marginal regression example | MSE: 0.52 -> 0.58 | Hurt |
| Refit or bias-correct after validation rather than using the naive validation estimator | Ridge regression · Comparison of cross-validation methods | accuracy | Helped |
| Choose λ by the one-standard-error rule | Ridge regression · Cross-validation simulations | test error | Hurt |
| Use Gaussian dual sketching with an initially suboptimal regularization parameter | Ridge regression · Gaussian dual sketching | MSE | Helped |
Closest papers by meaning
- Risk and cross validation in ridge regression with correlated samples (ICML 2025 · similarity 0.90)
- Asymptotically Free Sketched Ridge Ensembles: Risks, Cross-Validation, and Tuning (ICLR 2024 · similarity 0.90)
- Spurious Correlations in High Dimensional Regression: The Roles of Regularization, Simplicity Bias and Over-Parameterization (ICML 2025 · similarity 0.87)
- Optimal Ridge Regularization for Out-of-Distribution Prediction (ICML 2024 · similarity 0.87)
- An Agnostic View on the Cost of Overfitting in (Kernel) Ridge Regression (ICLR 2024 · similarity 0.87)
- Pooling Versus Ensembling for Ridge Regression Under Covariate Shift (NeurIPS 2026 · similarity 0.86)
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.