Term of the Moment
data
Definition: GAN
(1) For the alloy, see gallium nitride.
(2) For the GSM standard, see generic access network.
(3) (Generative Adversarial Network) An AI neural network that creates realistic images to train machine learning systems. To develop its expertise, thousands of real images are fed into a GAN from which the variations are created. GANs are also used to develop product alternatives and specific personas for advertisements. They can generate new video frames that turn a cloudy day into sunshine.
A GAN Is a Generator and Discriminator
The generator, which is a normal convolutional neural network (CNN), creates the images, and the discriminator (reverse CNN) evaluates them for realism. They work together to improve the quality of the images. As the discriminator provides feedback, the generator is able to fine-tune its output. See neural network.
Real and Generated
(2) For the GSM standard, see generic access network.
(3) (Generative Adversarial Network) An AI neural network that creates realistic images to train machine learning systems. To develop its expertise, thousands of real images are fed into a GAN from which the variations are created. GANs are also used to develop product alternatives and specific personas for advertisements. They can generate new video frames that turn a cloudy day into sunshine.
A GAN Is a Generator and Discriminator
The generator, which is a normal convolutional neural network (CNN), creates the images, and the discriminator (reverse CNN) evaluates them for realism. They work together to improve the quality of the images. As the discriminator provides feedback, the generator is able to fine-tune its output. See neural network.
Each pair of faces represents a real person and an image created in a generative adversarial network (GAN). (Image courtesy of NVIDIA Corporation.)