Research map / Optimization / Implicit bias and generalization
When is a Convolutional Filter Easy to Learn?
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 smoother input distribution (larger σ) | SGD for the synthetic one-neuron model · Synthetic one-neuron model | convergence speed | Helped |
| Make the convolutional patches closer (smaller σ₂) | SGD for a synthetic convolutional filter · Synthetic convolutional filter | convergence speed | Helped |
| Use a better initialization | SGD learning a random filter · MNIST | convergence speed | Helped |
| Use a better initialization | SGD learning a Gabor filter · MNIST | convergence speed | Helped |
Closest papers by meaning
- Globally Optimal Gradient Descent for a ConvNet with Gaussian Inputs (ICML 2017 · similarity 0.89)
- On Implicit Filter Level Sparsity in Convolutional Neural Networks (CVPR 2019 · similarity 0.89)
- CNN Filter DB: An Empirical Investigation of Trained Convolutional Filters (CVPR 2022 · similarity 0.88)
- RePr: Improved Training of Convolutional Filters (CVPR 2019 · similarity 0.88)
- Compressing Convolutional Neural Networks via Factorized Convolutional Filters (CVPR 2019 · similarity 0.88)
- A Convergence Theory for Deep Learning via Over-Parameterization (ICML 2019 · similarity 0.88)
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.