Research map / Label-efficient and robust learning / Recognition and long-tail
Revisiting Discriminative vs. Generative Classifiers: Theory and Implications
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 naïve Bayes instead of logistic regression as the linear evaluation classifier | Frozen pre-trained vision representations · CIFAR10 and CIFAR100 | convergence rate to asymptotic error: O(n) vs. O(log n) | Helped |
| Use naïve Bayes instead of logistic regression with limited training data | Supervised pre-trained vision representations · CIFAR10 and CIFAR100 | test error in few-shot regime | Mixed |
| Evaluate supervised rather than self-supervised pre-training | Pre-trained vision representations with naïve Bayes and logistic regression heads · CIFAR10 and CIFAR100 | occurrence of two regimes | Helped |
Closest papers by meaning
- Generative Classifiers Avoid Shortcut Solutions (ICLR 2025 · similarity 0.89)
- Intriguing Properties of Generative Classifiers (ICLR 2024 · similarity 0.88)
- Generative Cross-Entropy: A Strictly Proper Loss for Data-Efficient Classification (NeurIPS 2026 · similarity 0.86)
- Provably Improving Generalization of Few-shot models with Synthetic Data (ICML 2025 · similarity 0.85)
- SimPro: A Simple Probabilistic Framework Towards Realistic Long-Tailed Semi-Supervised Learning (ICML 2024 · similarity 0.85)
- On the Provable Advantage of Unsupervised Pretraining (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.