Research map / Architectures and efficiency / Neural architecture search
Off-Policy Reinforcement Learning for Efficient and Effective GAN Architecture Search
Yuan Tian, Qin Wang, Zhiwu Huang, Wen Li, Dengxin Dai, Minghao Yang, Jun Wang, Olga Fink
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add FID term (alpha=0.01) to the IS-only reward (ablation) | E2GAN · CIFAR-10 unconditional | FID: 15.64 -> 11.26 | Improved |
| add FID term (alpha=0.01) to the IS-only reward (ablation) | E2GAN · CIFAR-10 unconditional | Inception Score: 8.81 -> 8.51 | Hurt |
| add FID term to the reward (ablation) | E2GAN · CIFAR-10 unconditional | search cost (GPU days): 0.1 -> 0.3 | Hurt |
| replace AutoGAN on-policy multi-level search with off-policy MDP search (E2GAN) | GAN architecture search · CIFAR-10 unconditional | FID: 12.42 -> 11.26 | Improved |
| replace AutoGAN with E2GAN search | GAN architecture search · CIFAR-10 unconditional | Inception Score: 8.55 -> 8.51 | No clear effect |
| replace AutoGAN with E2GAN search | GAN architecture search · CIFAR-10 unconditional | search cost (GPU days): 2 -> 0.3 | Improved |
| replace AGAN with E2GAN search | GAN architecture search · CIFAR-10 unconditional | search cost (GPU days): 1200 -> 0.3 | Improved |
| use E2GAN-top2 instead of AutoGAN-top2 | GAN architecture search · CIFAR-10 unconditional | FID: 13.67 -> 12.96 | Improved |
| transfer E2GAN architecture instead of AutoGAN architecture to STL-10 | GAN architecture search · STL-10 unconditional | FID: 31.01 -> 25.35 | Improved |
| transfer E2GAN architecture instead of AutoGAN architecture to STL-10 | GAN architecture search · STL-10 unconditional | Inception Score: 9.16 -> 9.51 | Improved |
Closest papers by meaning
- EAGAN: Efficient Two-Stage Evolutionary Architecture Search for GANs (ECCV 2022 · similarity 0.90)
- AdversarialNAS: Adversarial Neural Architecture Search for GANs (CVPR 2020 · similarity 0.89)
- AutoGAN: Neural Architecture Search for Generative Adversarial Networks (ICCV 2019 · similarity 0.89)
- AlphaGAN: Fully Differentiable Architecture Search for Generative Adversarial Networks (TPAMI 2021 · similarity 0.88)
- Efficient Architecture Search by Network Transformation (AAAI 2018 · similarity 0.87)
- Towards Neural Architecture Search through Hierarchical Generative Modeling (ICML 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.
