Generative Adversarial Networks (GANs) are a class of machine learning models designed to automatically discover and learn the patterns in input data, creating new and novel data instances that resemble the input data distribution. They comprise two neural networks—a generator and a discriminator—competing in a zero-sum game, where the generator strives to create plausible data, and the discriminator attempts to discern real data from fake, refining both networks over time.