# AGAN: Towards Automated Design of Generative Adversarial Networks

arXiv: https://arxiv.org/abs/1906.11080 · HTML: https://mlsanity.com/paper/1906.11080/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Jun 2019.

## Built on by (most cited first, 8 of 8)

- [Neural Architecture Search: Insights from 1000 Papers](https://mlsanity.com/paper/2301.08727/) (2023)
- [Partition-Guided GANs](https://mlsanity.com/paper/2104.00816/) (2021)
- [Dynamically Grown Generative Adversarial Networks](https://mlsanity.com/paper/2106.08505/) (2021)
- [EGANS: Evolutionary Generative Adversarial Network Search for Zero-Shot Learning](https://mlsanity.com/paper/2308.09915/) (2023)
- [EAGAN: Efficient Two-stage Evolutionary Architecture Search for GANs](https://mlsanity.com/paper/2111.15097/) (2021)
- [Network representation learning: A macro and micro view](https://mlsanity.com/paper/2111.10772/) (2021)
- [CycleGANAS: Differentiable Neural Architecture Search for CycleGAN](https://mlsanity.com/paper/2311.07162/) (2023)
- [Combining Optimal Path Search With Task-Dependent Learning in a Neural Network](https://mlsanity.com/paper/2201.11104/) (2022)

## Abstract (arXiv)

Recent progress in Generative Adversarial Networks (GANs) has shown promising signs of improving GAN training via architectural change. Despite some early success, at present the design of GAN architectures requires human expertise, laborious trial-and-error testings, and often draws inspiration from its image classification counterpart. In the current paper, we present the first neural architecture search algorithm, automated neural architecture search for deep generative models, or AGAN for abbreviation, that is specifically suited for GAN training. For unsupervised image generation tasks on CIFAR-10, our algorithm finds architecture that outperforms state-of-the-art models under same regularization techniques. For supervised tasks, the automatically searched architectures also achieve highly competitive performance, outperforming best human-invented architectures at resolution 32×32. Moreover, we empirically demonstrate that the modules learned by AGAN are transferable to other image generation tasks such as STL-10.