Research map / Learning theory and algorithms / Kernels and regression theory
Bayes beats Cross Validation: Efficient and Accurate Ridge Regression via Expectation Maximization
Shu Tew, Mario Boley, Daniel F. Schmidt
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 |
|---|---|---|---|
| Tune lambda with Bayesian EM instead of fast LOOCV | Ridge regression hyperparameter tuning · UCI/UCR linear regression datasets | runtime ratio: up to 49 times faster | Improved |
| Use EM instead of fast LOOCV on Crop time series | Ridge regression hyperparameter tuning · Crop (multitarget) | speedup: 49 | Improved |
| Use EM instead of fast LOOCV on very small n (F.Fires, 2nd order) | Ridge regression hyperparameter tuning · F.Fires | speedup: 0.5 | Hurt |
| Use EM instead of fixed-grid LOOCV on StarL | Ridge regression hyperparameter tuning · StarL | accuracy: 0.60 -> 0.98 | Improved |
| Use EM instead of fixed-grid LOOCV on Eye | Ridge regression hyperparameter tuning · Eye linear | R2: 0.26 -> 0.50 | Improved |
| Use EM instead of fixed-grid LOOCV on Crime 2nd order | Ridge regression hyperparameter tuning · Crime | R2: -0.74 -> 0.67 | Improved |
| Use EM instead of fixed-grid LOOCV on A.mobile 2nd order | Ridge regression hyperparameter tuning · A.mobile | R2: -4e3 -> 0.90 | Improved |
| Use EM instead of glmnet-grid LOOCV on Eye linear | Ridge regression hyperparameter tuning · Eye | R2: 0.45 -> 0.50 | Improved |
| Use EM instead of glmnet-grid LOOCV on PT-motor 3rd order | Ridge regression hyperparameter tuning · PT-motor | R2: 0.04 -> -1.09 | Hurt |
| Use EM instead of glmnet-grid LOOCV on Abalone 2nd order | Ridge regression hyperparameter tuning · Abalone | R2: 0.50 -> 0.38 | Hurt |
| Use EM instead of LOOCV on most standard datasets (linear) | Ridge regression hyperparameter tuning · UCI datasets linear | R2: 0.94 vs 0.94 | No clear effect |
| Use EM instead of glmnet-grid LOOCV on Crop | Ridge regression hyperparameter tuning · Crop | accuracy: 0.76 -> 0.75 | No clear effect |
Closest papers by meaning
- Ridge Regression: Structure, Cross-Validation, and Sketching (ICLR 2020 · similarity 0.91)
- Risk and cross validation in ridge regression with correlated samples (ICML 2025 · similarity 0.90)
- Benign overfitting in ridge regression (JMLR 2023 · similarity 0.89)
- Optimal Ridge Regularization for Out-of-Distribution Prediction (ICML 2024 · similarity 0.88)
- Asymptotically Free Sketched Ridge Ensembles: Risks, Cross-Validation, and Tuning (ICLR 2024 · similarity 0.87)
- An Agnostic View on the Cost of Overfitting in (Kernel) Ridge Regression (ICLR 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.
