Research map / Learning theory and algorithms / Kernels and regression theory
Benign Overfitting in Multiclass Classification: All Roads Lead to Interpolation
Ke Wang, Vidya Muthukumar, Christos Thrampoulidis
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 |
|---|---|---|---|
| train with cross-entropy (multiclass SVM) instead of squared loss (MNI) under sufficient overparameterization | linear multiclass classifier, GMM data, k=4, n=40 · synthetic Gaussian mixture, p from 50 to 1200 | classification error | No clear effect |
| fewer classes k at same sample size n | multiclass SVM, MLM data, p=1000 · synthetic multinomial logistic model | fraction of support vectors satisfying interpolation constraint | Improved |
Closest papers by meaning
- The Implicit Bias of Benign Overfitting (JMLR 2023 · similarity 0.90)
- Generalization for multiclass classification with overparameterized linear models (NeurIPS 2022 · similarity 0.89)
- Risk Bounds for Over-parameterized Maximum Margin Classification on Sub-Gaussian Mixtures (NeurIPS 2021 · similarity 0.89)
- Rethinking Benign Overfitting in Two-Layer Neural Networks (ICML 2025 · similarity 0.88)
- Transfer Learning for Benign Overfitting in High-Dimensional Linear Regression (NeurIPS 2025 · similarity 0.87)
- Beyond Benign Overfitting in Nadaraya-Watson Interpolators (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.
