Ten Years of Generative Adversarial Nets (GANs): A survey of the state-of-the-art.
Machine Learning: Science and Technology., 2024
Generative Adversarial Networks (GANs) have become a cornerstone of modern generative modeling, enabling the creation of realistic and diverse data across domains. This survey provides a comprehensive overview of major GAN architectures, their theoretical foundations, and evaluation metrics. It also reviews key advancements, training challenges, and emerging integrations with frameworks such as Transformers and diffusion models.
Timeline of the application-based GAN architectures reviewed in this study. In this figure, CV indicates the Computer Vision domain and NLP indicates the field of Natural Language Processing.