skip to content

Search
← All concepts

#gan

Adversarial generator/discriminator training.

5 chapters, oldest era first: one idea traced across the books.

  1. From Noise to Image 2014 Ch. 3: The Adversarial Game The GAN minimax objective, its JS-divergence interpretation, and the failure mode called mode collapse.
  2. From Noise to Image 2015–2017 Ch. 4: Taming GAN Training DCGAN recipes, the Wasserstein distance, gradient penalties, and the birth of quantitative image-quality metrics.
  3. From Noise to Image 2015–2018 Ch. 7: First Words to Pictures Text conditioning before diffusion: StackGAN, AttnGAN, and attention over words.
  4. From Frames to Worlds 2016–2019 Ch. 2: GAN Video Scene dynamics, content–motion decomposition, and DVD-GAN's attempt to scale adversarial video.
  5. From Noise to Image 2017–2020 Ch. 6: The GAN Golden Age ProGAN to StyleGAN2: style modulation, latent arithmetic, and photorealistic faces of people who do not exist.