Research map / Image generation / GANs
Group Equivariant Generative Adversarial 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 standard convolutions with p4m group-convolutions in both networks (ablation) | Conditional GAN · ANHIR | FID: 7.32 -> 5.54 | Helped |
| Use p4m group-convolutions in both networks (ablation) | Conditional GAN · LYSTO | FID: 7.27 -> 3.90 | Helped |
| Use p4 group-convolutions in the discriminator only (ablation) | Conditional GAN · CIFAR-10 | FID: 20.89 -> 17.09 | Helped |
| Use p4 group-convolutions in both networks instead of discriminator only (ablation) | Conditional GAN · CIFAR-10 | FID: 17.09 -> 17.49 | Hurt |
| Use p4 group-convolutions in the discriminator only (ablation) | Conditional GAN · Food-101 | FID: 27.34 -> 16.91 | Helped |
| Use p4 group-convolutions in both networks rather than the discriminator only (ablation) | Conditional GAN · Food-101 | FID: 16.91 -> 17.73 | Hurt |
| Combine equivariant G and D with bCR (ablation) | Conditional GAN · ANHIR | FID: 5.54 -> 5.19 | Helped |
| Combine equivariant G and D with bCR (ablation) | Conditional GAN · LYSTO | FID: 3.90 -> 4.53 | Hurt |
| Combine equivariant G and D with bCR (ablation) | Conditional GAN · CIFAR-10 | FID: 17.49 -> 17.94 | Hurt |
| Combine equivariant G and D with bCR (ablation) | Conditional GAN · Food-101 | FID: 17.73 -> 15.55 | Helped |
| Use p4 group-convolutions in both networks (ablation) | Conditional GAN · Rotated MNIST, 33% training data, NSGAN | minimum FID: 0.74 -> 1.04 | Hurt |
| Use p4 group-convolutions in the discriminator only (ablation) | Conditional GAN · Rotated MNIST, 10% training data, RaGAN | mean FID: 11.40 -> 4.26 | Helped |
Closest papers by meaning
- Variational Partial Group Convolutions for Input-Aware Partial Equivariance of Rotations and Color-Shifts (ICML 2024 · similarity 0.90)
- Structure-preserving GANs (ICML 2022 · similarity 0.90)
- Probing Equivariance and Symmetry Breaking in Convolutional Networks (NeurIPS 2025 · similarity 0.90)
- Learning Color Equivariant Representations (ICLR 2025 · similarity 0.88)
- Lie Group Decompositions for Equivariant Neural Networks (ICLR 2024 · similarity 0.88)
- Bridging Symmetry and Robustness: On the Role of Equivariance in Enhancing Adversarial Robustness (NeurIPS 2025 · 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.