Also known as: gan Β· gans Β· Generative Adversarial Network (GAN) Β· Generative Adversarial Networks Β· GAN
Generative Adversarial Network (GAN) is a machine learning architecture in which a generator and discriminator are trained in opposition. The generator learns to create realistic samples, while the discriminator learns to distinguish generated samples from real data.
A class of machine learning systems. Two neural networks contest with each other in a zero-sum game framework.
A GAN is a type of artificial intelligence framework involving two neural networks, typically referred to as the generator and the discriminator, which are trained simultaneously through a process of competition. The generator networks create new data that resemble the training data β it starts with random noise and transforms it into data, such as images, that mimic the real dataset. The discriminator evaluates the data it receives, distinguishing between real data (from the training set) and fake data (produced by the generator), acting like a judge that decides whether a given instance is authentic or fake.
A system to create new data in which a generator creates data and a discriminator determines whether that created data is valid or invalid. See the Generative Adversarial Networks course for more information.