GANs are a type of machine learning algorithm that involves defining a problem as a supervised learning problem with two sub-models.
The AI model is trained to create a new set of data points belonging to a particular domain. In contrast, the classifier model, called a discriminator, identifies the new set of data points as real or fake. In this type of repetitive training, the generator takes the risk of generating examples closer to reality, while the discriminator becomes wiser in determining which samples are fake and real.
