AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Towards Diffeomorphism-Equivariant Neural Networks via Canonicalization

Josephine Elisabeth Oettinger, Zakhar Shumaylov, Johannes Bostelmann, Jan Lellmann, Carola-Bibiane Schönlieb

ICML 2026 · Architectures and efficiency › CNN design · approach: Group symmetry equivariant networks

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
add energy canonicalization (DiffeoNN) in front of a pre-trained U-NetU-Net segmentation · Synthetic nested squares, transformed testIoU: 0.9276 -> 0.9571Improved
add energy canonicalization (DiffeoNN) in front of a pre-trained U-NetU-Net segmentation · Lung segmentation, transformed chest X-raysIoU: 0.9147 -> 0.9214Improved
add energy canonicalization to a CNN genus classifierMNIST genus CNN · Transformed MNISTaccuracy: 0.68 -> 0.82Improved
train on diffeomorphic data augmentation instead of canonicalizationU-Net segmentation · Synthetic nested squaresIoU: 0.9571 -> 0.9770Improved
use canonicalization instead of augmentation (fewer outliers)U-Net segmentation · Synthetic and lung segmentationDice outliersImproved
use VAE energy with or without adversarial energy instead of no canonicalization (ablation)DiffeoNN canonicalization energy · Lung segmentation, 100 transformed X-raysIoUImproved
combine adversarial, VAE and regularization energy terms (ablation)DiffeoNN canonicalization energy · Lung segmentation, 100 transformed X-raysIoUImproved
use adversarial energy alone or with only regularization (ablation)DiffeoNN canonicalization energy · Lung segmentation, 100 transformed X-raysIoUHurt

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