AI Paper Atlas

Research map / Architectures and efficiency / CNN design

On the Ability of Deep Networks to Learn Symmetries from Data: A Neural Kernel Theory

Andrea Perin, Stephane Deny

JMLR 2025 · Architectures and efficiency › CNN design · approach: Group symmetry equivariant networks · 6 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (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 changedWhereMeasuredEffect
increase the number of rotation angles in the orbit of rotated-MNISTMLP, ConvNet and ViT-S · rotated-MNIST (leave-out class, out-test)out-test accuracyImproved
train for very long times (over 1000 epochs)MLP, ConvNet and ViT-S · rotated-MNIST (leave-out class, out-test)out-test accuracyNo clear effect
use a global average pooling convolutional network instead of a fully connected convolutional networkconvolutional network · translated-MNIST (pairs of orbits)error on left-out pointImproved
use non-overlapping filters (stride equals filter size) in the GAP convolutional network (ablation)GAP convolutional network · rotated-MNIST (pairs of orbits)orbit classification errorHurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy