Research map / Architectures and efficiency / CNN design
What Can Be Learnt With Wide Convolutional Neural Networks?
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 a hierarchical CNN kernel rather than a fully-connected kernel | Kernel regression on a target depending on a local meta-patch · Localized target functions | generalisation error decay exponent β: β = 2m(2ν+d_eff(L))/(2m(2ν+d_eff(L))+(2ν+d_eff(l))d_eff(L)) vs β = 2m/(2m + p) | Helped |
| Use a hierarchical CNN kernel on a target depending on the full input | Hierarchical CNN kernel regression · Global target functions | generalisation error decay exponent β: β = 3/d_eff(L) | Hurt |
| Use a modified LeNet with nonoverlapping patches and no pooling instead of standard LeNet | LeNet on CIFAR10 · CIFAR10 test | accuracy: 62% -> 53% | Hurt |
| Train a hierarchical CNN kernel on the output of a randomly-initialised infinitely-wide hierarchical network | Hierarchical CNN kernel regression · Hierarchical teacher targets | optimal generalisation error decay exponent β: β = 3/d_eff(L) | Hurt |
Closest papers by meaning
- Minimax optimality of convolutional neural networks for infinite dimensional input-output problems and separation from kernel methods (ICLR 2024 · similarity 0.90)
- Disentangling Trainability and Generalization in Deep Neural Networks (ICML 2020 · similarity 0.88)
- The Computational Advantage of Depth in Learning High-Dimensional Hierarchical Targets (NeurIPS 2025 · similarity 0.88)
- Optimization Landscape and Expressivity of Deep CNNs (ICML 2018 · similarity 0.88)
- Linear CNNs Discover the Statistical Structure of the Dataset Using Only the Most Dominant Frequencies (ICML 2023 · similarity 0.88)
- How Deep Networks Learn Sparse and Hierarchical Data: the Sparse Random Hierarchy Model (ICML 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.