Research map / Learning theory and algorithms / Sample complexity
Few-Shot Learning via Learning the Representation, Provably
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 |
|---|---|---|---|
| Learn a shared low-dimensional linear representation from source tasks | Linear regression on the target task · Target task | sample complexity: O(d) -> O(k) | Helped |
| Pool source-task samples to learn a shared representation | Low-dimensional linear representation learning · Target task | expected excess risk: \sigma^2 \left( \frac{kd \log(\kappa n_1)}{cn_1T} + \frac{k+\log\frac1\delta}{n_2} \right) vs. O\left(\frac{d}{n_2}\righ | Helped |
| Learn a norm-controlled shared linear representation from source tasks | High-dimensional linear regression · Target task | expected excess risk: \sigma \bar R \cdot \tilde O\left(\frac{\sqrt{\Trace(\Sigma)}}{\sqrt{n_1T}}+\frac{\sqrt{\|\Sigma\|_2}}{\sqrt{n_2}} \righ | Helped |
| Learn a shared representation with a two-layer ReLU network from source tasks | Two-layer ReLU network on the target task · Target task | expected excess risk: \sigma \bar R \cdot \tilde O\left(\frac{\sqrt{\Trace(\Sigma)}}{\sqrt{n_1T}}+\frac{\sqrt{\|\Sigma\|_2}}{\sqrt{n_2}} \righ | Helped |
Closest papers by meaning
- Guarantees for Nonlinear Representation Learning: Non-identical Covariates, Dependent Data, Fewer Samples (ICML 2024 · similarity 0.88)
- Sample-Efficient Linear Representation Learning from Non-IID Non-Isotropic Data (ICLR 2024 · similarity 0.88)
- BECLR: Batch Enhanced Contrastive Few-Shot Learning (ICLR 2024 · similarity 0.87)
- Provable Sample-Efficient Transfer Learning Conditional Diffusion Models via Representation Learning (NeurIPS 2025 · similarity 0.86)
- Optimal Representation Size: High-Dimensional Analysis of Pretraining and Linear Probing (NeurIPS 2026 · similarity 0.86)
- Flatten Long-Range Loss Landscapes for Cross-Domain Few-Shot Learning (CVPR 2024 · 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.