Addressing GAN Training Instabilities via Tunable Classification Losses

Understand

Generative adversarial networks (GANs), modeled as a zero-sum game between a generator (G) and a discriminator (D), allow generating synthetic data with formal guarantees.

• Noting that D is a classifier, we begin by reformulating the GAN value function using class probability estimation (CPE) losses.
• We prove a two-way correspondence between CPE loss GANs and $f$-GANs which minimize $f$-divergences.
• We also show that all symmetric $f$-divergences are equivalent in convergence.