Research map / Learning theory and algorithms / Kernels and regression theory
Exact expressions for double descent and implicit regularization via surrogate random design
Michal Derezinski, Feynman T Liang, Michael W. Mahoney
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 |
|---|---|---|---|
| Use a determinant-weighted surrogate random design to derive MSE expressions | Moore-Penrose estimator · Linear regression under surrogate design | MSE characterization | Improved |
| Use the minimum-norm estimator on an under-determined surrogate sample | Moore-Penrose estimator · Surrogate design, n<d | Expected estimator | Improved |
| Vary the sample size across the interpolation threshold | Moore-Penrose estimator · Linear regression under surrogate design | MSE | Mixed |
| Use covariance matrices with different spectral decay | Moore-Penrose estimator · Under-determined linear regression | MSE | Mixed |
| Use rapidly decaying covariance eigenvalues | Moore-Penrose estimator · Under-determined linear regression | MSE relative to the null estimator | Improved |
| Use the surrogate MSE expression when the signal-to-noise ratio is 1 | Moore-Penrose estimator · Under-determined linear regression with non-isotropic covariance | MSE | Improved |
Closest papers by meaning
- Overfitting Can Be Harmless for Basis Pursuit, But Only to a Degree (NeurIPS 2020 · similarity 0.88)
- Implicit Sparse Regularization: The Impact of Depth and Early Stopping (NeurIPS 2021 · similarity 0.88)
- Optimal Regularization can Mitigate Double Descent (ICLR 2021 · similarity 0.88)
- On the Role of Optimization in Double Descent: A Least Squares Study (NeurIPS 2021 · similarity 0.88)
- Unraveling the Enigma of Double Descent: An In-depth Analysis through the Lens of Learned Feature Space (ICLR 2024 · similarity 0.86)
- Implicit Bias of Spectral Descent and Muon on Multiclass Separable Data (NeurIPS 2025 · 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.
