Research map / Architectures and efficiency / CNN design
A General Framework for Robust G-Invariance in G-Equivariant 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 |
|---|---|---|---|
| Replace Max G-Pooling with G-TC layer | G-CNN · SO(2)-MNIST (C8-CNN) | test accuracy (%): 95.23 -> 96.53 | Helped |
| Replace Max G-Pooling with G-TC layer | G-CNN · O(2)-MNIST (D16-CNN) | test accuracy (%): 92.17 -> 93.06 | Helped |
| Replace Max G-Pooling with G-TC layer | G-CNN · O-ModelNet10 | test accuracy (%): 72.17 -> 74.01 | Helped |
| Replace Max G-Pooling with G-TC layer | G-CNN · Oh-ModelNet10 | test accuracy (%): 71.73 -> 75.22 | Helped |
| Replace Max G-Pooling with G-TC layer | G-CNN · Oh-ModelNet10 params | parameters: 1,826,978 -> 1,817,602 | Helped |
| Replace Max G-Pooling with G-TC layer | G-CNN · O-ModelNet10 params | parameters: 500,198 -> 472,066 | Helped |
| Replace Max G-Pooling with G-TC layer | G-CNN · SO(2)-MNIST params | parameters: 32,915 -> 35,218 | No clear effect |
| Replace Max G-Pooling with G-TC layer | G-CNN · MNIST metamer optimization | metamer completeness | Helped |
| Replace Max G-Pooling with G-TC layer | G-CNN · O-ModelNet10 seed variance | accuracy std: 0.95 -> 0.48 | Helped |
| Replace Max G-Pooling with G-TC layer | G-CNN · Oh-ModelNet10 seed variance | accuracy std: 0.23 -> 0.62 | Hurt |
| Remove nonlinearity from G-Conv block when using G-TC (third-order TC nonlinearity) | G-CNN · G-MNIST/ModelNet10 | design choice | No clear effect |
Closest papers by meaning
- Variational Partial Group Convolutions for Input-Aware Partial Equivariance of Rotations and Color-Shifts (ICML 2024 · similarity 0.88)
- Group Downsampling with Equivariant Anti-aliasing (ICLR 2025 · similarity 0.87)
- Bridging Symmetry and Robustness: On the Role of Equivariance in Enhancing Adversarial Robustness (NeurIPS 2025 · similarity 0.87)
- Equivariance by Contrast: Identifiable Equivariant Embeddings from Unlabeled Finite Group Actions (NeurIPS 2025 · similarity 0.86)
- Probing Equivariance and Symmetry Breaking in Convolutional Networks (NeurIPS 2025 · similarity 0.86)
- Lie Group Decompositions for Equivariant Neural Networks (ICLR 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.