Research map / Image generation / GANs
Adaptive Weighted Discriminator for Training 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 |
|---|---|---|---|
| Replace equally weighted discriminator loss with adaptive gradient-based weights | AutoGAN · CIFAR-10 unconditional | IS: 8.55{ $\pm$.10} -> 9.01{ $\pm$.03} | Helped |
| Replace equally weighted discriminator loss with adaptive gradient-based weights | AutoGAN · CIFAR-10 unconditional | FID: 12.42 -> 11.82 | Helped |
| Replace equally weighted discriminator loss with adaptive gradient-based weights | AutoGAN · STL-10 unconditional | IS: 9.16{ $\pm$.12} -> 9.41{ $\pm$.09} | Helped |
| Replace equally weighted discriminator loss with adaptive gradient-based weights | AutoGAN · STL-10 unconditional | FID: 31.01 -> 26.32 | Helped |
| Replace equally weighted discriminator loss with adaptive gradient-based weights | AutoGAN · CIFAR-100 unconditional | IS: 8.54{ $\pm$.10} -> 8.90{ $\pm$.06} | Helped |
| Replace equally weighted discriminator loss with adaptive gradient-based weights | AutoGAN · CIFAR-100 unconditional | FID: 19.98$^{*}$ -> 19.00 | Helped |
| Use adaptive weighted discriminator loss | BigGAN · CIFAR-10 conditional | FID: 14.73 -> 7.03 | Helped |
| Use adaptive weighted discriminator loss | BigGAN · CIFAR-100 conditional | FID: 11.73 -> 10.23 | Helped |
| Use non-normalized rather than normalized gradients (ablation) | aw-AutoGAN · CIFAR-10 unconditional | IS: 9.01{ $\pm$.03} -> 8.98{ $\pm$.06} | Hurt |
| Use non-normalized rather than normalized gradients (ablation) | aw-AutoGAN · CIFAR-10 unconditional | FID: 11.82 -> 13.17 | Hurt |
| Use non-normalized rather than normalized gradients (ablation) | aw-BigGAN · CIFAR-10 conditional | FID: 7.03 -> 6.89 | Helped |
| Use non-normalized rather than normalized gradients (ablation) | aw-BigGAN · CIFAR-100 conditional | IS: 11.22{ $\pm$.17} -> 11.26{ $\pm$.20} | Helped |
Closest papers by meaning
- Improving Generalization and Stability of Generative Adversarial Networks (ICLR 2019 · similarity 0.89)
- Training Generative Adversarial Networks with Limited Data (NeurIPS 2020 · similarity 0.89)
- Dual Discriminator Generative Adversarial Nets (NeurIPS 2017 · similarity 0.89)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (AAAI 2025 · similarity 0.88)
- Improving the Training of Data-Efficient GANs via Quality Aware Dynamic Discriminator Rejection Sampling (CVPR 2025 · similarity 0.87)
- Discriminative Forests Improve Generative Diversity for Generative Adversarial Networks (AAAI 2024 · similarity 0.87)
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.