Research map / Learning theory and algorithms / Sample complexity
Towards Bridging Sample Complexity and Model Capacity
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 |
|---|---|---|---|
| Add cutout augmentation | CIFAR10 original images · CIFAR10 | Clusters: 493.2 -> 730.0 | Helped |
| Add cutout augmentation | CIFAR10 original images · CIFAR10 | decision tree nodes: 4993 -> 5303 | Helped |
| Add random crop augmentation | CIFAR10 original images · CIFAR10 | Clusters: 493.2 -> 636.1 | Helped |
| Add affine transformation augmentation | CIFAR10 original images · CIFAR10 | Clusters: 493.2 -> 564.3 | Helped |
| Add rotation augmentation | CIFAR10 original images · CIFAR10 | Clusters: 493.2 -> 579.6 | Helped |
| Merge adjacent category pairs into 5 classes | CIFAR10 data combinations · CIFAR10 | decision tree nodes: 23 -> 67 | Helped |
| Merge similar category pairs into 5 classes | CIFAR10 data combinations · CIFAR10 | decision tree nodes: 27 -> 65 | Helped |
| Apply PGD adversarial attack | CIFAR10 images · CIFAR10 adversarial examples | decision tree nodes: 99 -> 1835 | Helped |
| Apply PGD adversarial attack | CIFAR10 images · CIFAR10 adversarial examples | test accuracy: 92.15 -> 47.05 | Hurt |
| Apply PGD adversarial attack | CIFAR10 images · CIFAR10 adversarial examples | distribution overlapping: 0.296 -> 37.791 | Helped |
| Apply PGD adversarial attack | CIFAR10 images · CIFAR10 adversarial examples | distribution dispersity: 0.592 -> 6.889 | Helped |
Closest papers by meaning
- Optimal Sample Complexity of Contrastive Learning (ICLR 2024 · similarity 0.87)
- Provable Sample-Efficient Transfer Learning Conditional Diffusion Models via Representation Learning (NeurIPS 2025 · similarity 0.85)
- Nearly Optimal Sample Complexity for Learning with Label Proportions (ICML 2025 · similarity 0.85)
- Improved Sample Complexity Bounds for Diffusion Model Training (NeurIPS 2024 · similarity 0.85)
- Sample Complexity and Representation Ability of Test-time Scaling Paradigms (ICLR 2026 · similarity 0.84)
- Guarantees for Nonlinear Representation Learning: Non-identical Covariates, Dependent Data, Fewer Samples (ICML 2024 · similarity 0.84)
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.