Adversarial Generative Models
The minimax game, mode collapse, Wasserstein critics, and what FID does and does not measure.
7concepts
76flashcards
53minutes of reading
- 01 Mode Collapse and Training Instability Why a GAN generator has no incentive to cover the data distribution, the difference between full and partial collapse, and what each of the standard mitigations actually changes.
- 02 StyleGAN and Style-Based Generation How StyleGAN moved the latent code out of the generator's input and into per-layer modulation through a learned mapping network, and how StyleGAN2 and StyleGAN3 traced visible artefacts back to normalisation, progressive growing and aliasing.
- 03 Wasserstein Critics and the Lipschitz Constraint Why earth mover distance still has a gradient when JS divergence does not, how the Kantorovich duality turns it into a trainable critic, and why enforcing the Lipschitz bound is where every practical difficulty lives.