# A New Perspective on Stabilizing GANs training: Direct Adversarial Training

arXiv: https://arxiv.org/abs/2008.09041 · HTML: https://mlsanity.com/paper/2008.09041/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Aug 2020.

## Built on by (most cited first, 12 of 12)

- [Is Artificial Intelligence Generated Image Detection a Solved Problem?](https://mlsanity.com/paper/2505.12335/) (2025)
- [Domain Re-Modulation for Few-Shot Generative Domain Adaptation](https://mlsanity.com/paper/2302.02550/) (2023)
- [Exploring The Effect of High-frequency Components in GANs Training](https://mlsanity.com/paper/2103.11093/) (2021)
- [Pedestrian Trajectory Prediction Based on Social Interactions Learning With Random Weights](https://mlsanity.com/paper/2501.07711/) (2025)
- [Efficient Backdoor Attacks for Deep Neural Networks in Real-world Scenarios](https://mlsanity.com/paper/2306.08386/) (2023)
- [A Proxy Attack-Free Strategy for Practically Improving the Poisoning Efficiency in Backdoor Attacks](https://mlsanity.com/paper/2306.08313/) (2023)
- [Peer is Your Pillar: A Data-unbalanced Conditional GANs for Few-shot Image Generation](https://mlsanity.com/paper/2311.08217/) (2023)
- [One-shot Generative Domain Adaptation in 3D GANs](https://mlsanity.com/paper/2410.08824/) (2024)
- [Efficient Trigger Word Insertion](https://mlsanity.com/paper/2311.13957/) (2023)
- [Large Language Models are Good Attackers: Efficient and Stealthy Textual Backdoor Attacks](https://mlsanity.com/paper/2408.11587/) (2024)
- [Leveraging Generalizability of Image-to-Image Translation for Enhanced Adversarial Defense](https://mlsanity.com/paper/2504.01399/) (2025)
- [MM2CT: MR-to-CT translation for multi-modal image fusion with mamba](https://mlsanity.com/paper/2508.05476/) (2025)

## Abstract (arXiv)

Generative Adversarial Networks (GANs) are the most popular image generation models that have achieved remarkable progress on various computer vision tasks. However, training instability is still one of the open problems for all GAN-based algorithms. Quite a number of methods have been proposed to stabilize the training of GANs, the focuses of which were respectively put on the loss functions, regularization and normalization technologies, training algorithms, and model architectures. Different from the above methods, in this paper, a new perspective on stabilizing GANs training is presented. It is found that sometimes the images produced by the generator act like adversarial examples of the discriminator during the training process, which may be part of the reason causing the unstable training of GANs. With this finding, we propose the Direct Adversarial Training (DAT) method to stabilize the training process of GANs. Furthermore, we prove that the DAT method is able to minimize the Lipschitz constant of the discriminator adaptively. The advanced performance of DAT is verified on multiple loss functions, network architectures, hyper-parameters, and datasets. Specifically, DAT achieves significant improvements of 11.5% FID on CIFAR-100 unconditional generation based on SSGAN, 10.5% FID on STL-10 unconditional generation based on SSGAN, and 13.2% FID on LSUN-Bedroom unconditional generation based on SSGAN. Code will be available at https://github.com/iceli1007/DAT-GAN