Research map / Image generation / GANs
Stacked Generative Adversarial Networks
Check 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 |
|---|---|---|---|
| use a top-down stack of GANs with representation discriminators instead of a single GAN | SGAN · CIFAR-10 | Inception score: 7.16± 0.10 -> 8.59 ± 0.12 | Helped |
| add final end-to-end joint training (ablation) | SGAN · CIFAR-10 | Inception score: 8.37 ± 0.08 -> 8.59 ± 0.12 | Helped |
| add entropy loss to a conditional DCGAN (ablation) | DCGAN with adversarial and conditional losses · CIFAR-10 | Inception score: 5.40± 0.08 -> 7.16± 0.10 | Helped |
| add entropy loss without conditional loss (ablation) | DCGAN with adversarial loss · CIFAR-10 | Inception score: 6.16± 0.07 -> 5.40± 0.16 | Hurt |
| add conditional loss without entropy loss (ablation) | DCGAN with adversarial loss · CIFAR-10 | Inception score: 6.16± 0.07 -> 5.40± 0.08 | Hurt |
| add conditional and entropy losses together (ablation) | DCGAN with adversarial loss · CIFAR-10 | Inception score: 6.16± 0.07 -> 7.16± 0.10 | Helped |
| remove entropy loss (ablation) | conditional generator · MNIST, fixed fc3 feature | output diversity | Hurt |
| remove entropy loss (ablation) | conditional DCGAN · CIFAR-10 | output diversity | Hurt |
Closest papers by meaning
- SGAN: An Alternative Training of Generative Adversarial Networks (CVPR 2018 · similarity 0.90)
- Generative Multi-Adversarial Networks (ICLR 2017 · similarity 0.89)
- MGAN: Training Generative Adversarial Nets with Multiple Generators (ICLR 2018 · similarity 0.89)
- MSG-GAN: Multi-Scale Gradients for Generative Adversarial Networks (CVPR 2020 · similarity 0.89)
- Taming the Tail in Class-Conditional GANs: Knowledge Sharing via Unconditional Training at Lower Resolutions (CVPR 2024 · similarity 0.88)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 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.