Research · June 10, 2014

Generative adversarial networks

A training scheme was described in which a generator and a discriminator learn against each other.

Why it matters

GANs made realistic image generation a practical problem and set a research direction that preceded diffusion models.

Training such models is unstable, and much of the later work addresses exactly that problem.

Event date

June 10, 2014

Timeline date

Event date

Verification

Sources gathered automatically · September 16, 2026

Lines

ID

evt-0051

Related Diffusion models
Both describe image generation learned from data; the training mechanisms differ.

Contrasts The variational autoencoder
A different answer to the same generation problem, arriving six months earlier.
Auto-Encoding Variational Bayes
Contrasts Denoising diffusion probabilistic models
A different route to the same task: sample quality is measured against adversarial networks, without adversarial training.
Denoising Diffusion Probabilistic Models
Builds on pix2pix: one model for image translation
pix2pix is a conditional adversarial network: the generator learns to fool the discriminator, as in the GAN of 2014, which the paper cites.
Image-to-Image Translation with Conditional Adversarial Networks (arXiv:1611.07004v1)
Builds on StyleGAN: a style-controlled face generator
StyleGAN changes only the generator of an adversarial network, leaving the discriminator and the loss as in a GAN.
A Style-Based Generator Architecture for Generative Adversarial Networks (arXiv:1812.04948v1)