atlas

Research map / Architectures and efficiency / CNN design

Group Equivariant Capsule Networks

NeurIPS 2018 · Architectures and efficiency › CNN design · approach: Group symmetry equivariant networks · 93 citations (OpenAlex)

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 changedWhereMeasuredEffect
combine the capsule network with pose-guided convolutions (ablation)Group equivariant capsule network · MNIST rot. (50k)accuracy: 92.30% -> 98.42%Helped
combine the capsule network with pose-guided convolutions (ablation)Group equivariant capsule network · AffNistaccuracy: 81.64% -> 89.10%Helped
combine the capsule network with pose-guided convolutions (ablation)Group equivariant capsule network · MNIST rot. (10k)accuracy: 90.19% -> 97.40%Helped
add reconstruction loss (ablation)Capsule pose extraction · MNIST pose extractionaverage pose error [degree]: 28.32 -> 16.21Helped
use capsule pose routing instead of naive hierarchical pose averaging (ablation)MNIST pose extraction · MNIST pose extractionaverage pose error [degree]: 70.92 -> 28.32Helped

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