ConceptAI & ML
Generative adversarial networks
A generative model that learns to generate data from a training distribution through a two-player game between a generator and a discriminator network.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Opens up
Concepts that name this one as a prerequisite.
Lecturers
6Each in their own terms, from the easiest level.
Yoshua Bengio
Generative adversarial networks are models introduced around 2014 that can synthesize and imagine new realistic images based on training data.Phillip Isola
Generative models that exhibit oscillatory behavior and high-dimensional outputs during training.Sergey Levine
An approach to generative modeling that learns a neural network to capture a particular given data distribution using a generator and a discriminator network.Yann LeCun
A GAN is trained by having a second neural network tell the first neural network whether its prediction is on the plausible data manifold.Sara Beery
A type of generative model that can be used to explain classifiers by exploring counterfactuals in the latent space.
Lectures that assume it
1Connections
18The lectures that define it or assume it, and the concepts next to it.