Research map / Learning theory and algorithms / Kernels and regression theory
On Uniform Convergence and Low-Norm Interpolation Learning
Lijia Zhou, Danica J. Sutherland, Nati Srebro
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 |
|---|---|---|---|
| Restrict the norm-bounded class to predictors that interpolate the training data | Uniform convergence in junk-features Gaussian linear regression · Junk-features Gaussian linear regression | asymptotic worst-case excess risk: α^2 L_{\mathcal{D}}(w^*) | Improved |
| Allow interpolators norm at most α times the minimum norm, with α ≥ 1 | Minimum-norm interpolation · Junk-features Gaussian linear regression | asymptotic worst-case excess risk: α^2 L_{\mathcal{D}}(w^*) | Mixed |
| Use a norm ball containing the minimum-norm interpolator without requiring interpolation | Minimum-norm interpolator · Junk-features Gaussian linear regression | two-sided uniform generalization gap: \infty | Hurt |
| Use typical algorithm outputs in a two-sided uniform convergence bound | Symmetric interpolation algorithms · Junk-features Gaussian linear regression | two-sided uniform generalization gap: 3 \sigma^2 | Hurt |
| Set the ridge parameter to λ_n = o(n) | Minimum-norm interpolator · Junk-features Gaussian linear regression | expected excess population risk: 0 | Improved |
| Use an interpolator with norm no more than a constant factor above the minimum norm | Low-norm interpolation · Junk-features Gaussian linear regression | asymptotic excess population risk: α^2 L_{\mathcal{D}}(w^*) | Improved |
Closest papers by meaning
- Uniform Convergence of Interpolators: Gaussian Width, Norm Bounds and Benign Overfitting (NeurIPS 2021 · similarity 0.93)
- Failures of Model-dependent Generalization Bounds for Least-norm Interpolation (JMLR 2021 · similarity 0.90)
- Minimum Norm Interpolation Meets The Local Theory of Banach Spaces (ICML 2024 · similarity 0.88)
- Beyond Benign Overfitting in Nadaraya-Watson Interpolators (NeurIPS 2025 · similarity 0.86)
- Error Feedback under $(L_0,L_1)$-Smoothness: Normalization and Momentum (NeurIPS 2025 · similarity 0.85)
- Flat Minima in Linear Estimation and an Extended Gauss Markov Theorem (ICLR 2024 · similarity 0.85)
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.
