Abstract:Highlights•Generative Adversarial Networks (GANs) are notorious to train. Research has been conducted in various angles to address the challenges.•Classifiers or Discriminators are susceptible to adversarial perturbations. When trained with samples created by GANs, adversarial robustness of these models are enhanced.•Adversarial Training and GANs mutually help each other and exhibit potential to be considered for further research.Recognized as a realistic image generator, Generative Adversarial Network (GAN) occupies a progressive section in deep learning. Using generative modeling, the underlying generator model learns the real target distribution and outputs fake samples from the generated replica distribution. The discriminator attempts to distinguish the fake and the real samples and sends feedback to the generator so that the generator can improve the fake samples. Recently, GANs have been competing with the state-of-the-art in various tasks including image processing, missing data imputation, text-to-image translation and adversarial example generation. However, the architecture suffers from training instability, resulting in problems like non-convergence, mode collapse and vanishing gradients. The research community has been studying and devising modified architectures, alternative lost functions and techniques to address these concerns. A section of publications has studied Adversarial Training, alongside GANs. This review covers the existing works on the instability of GANs from square one and a portion of recent publications to illustrate the trend of research. It also gives insight to studies on adversarial attacks and research discussing Adversarial Attacks with GANs. To put it more eloquently, this study intends to guide researchers interested in studying improvisations made to GANs for stable training, in the presence of Adversarial Attacks.

Stabilizing Generative Adversarial Networks: A Survey

A Survey on Generative Adversarial Networks: Variants, Applications, and Training

Survey on Generative Adversarial Behavior in Artificial Neural Tasks

Generative Adversarial Networks (GANs Survey): Challenges, Solutions, and Future Directions

A New Perspective on Stabilizing GANs Training: Direct Adversarial Training

Why Adopting Regularization and Normalization For Generative Adversarial Networks: A Survey

Stabilizing Training of Generative Adversarial Networks through Regularization

A Survey of Modern Deep Learning based Generative Adversarial Networks (GANs)

Exploring generative adversarial networks and adversarial training

Challenges and Corresponding Solutions of Generative Adversarial Networks (GANs): A Survey Study

Smoothness and Stability in GANs

Understanding and Stabilizing GANs' Training Dynamics with Control Theory

Ten Years of Generative Adversarial Nets (GANs): A survey of the state-of-the-art

On Stabilizing Generative Adversarial Training with Noise

Training Generative Adversarial Networks by Solving Ordinary Differential Equations

Stabilizing Training of Generative Adversarial Nets via Langevin Stein Variational Gradient Descent

A Systematic Survey of Regularization and Normalization in GANs

A Chronological Survey of Theoretical Advancements in Generative Adversarial Networks for Computer Vision

Understanding GANs: fundamentals, variants, training challenges, applications, and open problems