Research map / Optimization / Implicit bias and generalization
The Performance Analysis of Generalized Margin Maximizers on Separable Data
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 ℓ2-GMM (max-margin) rather than ℓ1-GMM or ℓ∞-GMM | GMM classifiers with Gaussian underlying signals · Gaussian signals | generalization error | Helped |
| Use ℓ1-GMM rather than ℓ2-GMM or ℓ∞-GMM | GMM classifiers with sparse underlying signals · Sparse signals; δ approximately below 6 | generalization error | Helped |
| Use ℓ∞-GMM rather than ℓ2-GMM or ℓ1-GMM | GMM classifiers with binary underlying signals · Binary signals; δ<1 | generalization error | Helped |
| Use ℓ2-GMM rather than ℓ∞-GMM | GMM classifiers with binary underlying signals · Binary signals; larger δ | generalization error | Helped |
Closest papers by meaning
- Finite-Sample Performance of Gradient Descent in Logistic Regression with Gaussian Design (NeurIPS 2026 · similarity 0.87)
- Achieving Margin Maximization Exponentially Fast via Progressive Norm Rescaling (ICML 2024 · similarity 0.86)
- Gradient Descent Converges Arbitrarily Fast for Logistic Regression via Large and Adaptive Stepsizes (ICML 2025 · similarity 0.86)
- From Logistic Regression to the Perceptron Algorithm: Exploring Gradient Descent with Large Step Sizes (AAAI 2025 · similarity 0.85)
- Minimax Optimal Early-Stopped Gradient Descent for Gaussian Mixture Classification (NeurIPS 2026 · similarity 0.85)
- Multi-Class Support Vector Machine with Maximizing Minimum Margin (AAAI 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.