ml-recall

#cyclegan

2 questions from Generative Adversarial Networks

Timed test

2 questions · 2 min · score and explanations at the end

Take a test

Q 1 / 2Score 0 / 0

Shuffle questions

Question 1: Short answer

mediumAI-checkedGenerative Adversarial Networks

A CycleGAN discriminator is trained with the least-squares (LSGAN) loss: the real loss is the mean of $(D(x) - 1)^2$ over real images and the fake loss the mean of $D(G(z))^2$ over fakes; the discriminator loss is their sum. For two real images the outputs are 0.8 and 0.6, for two fakes 0.3 and 0.1. What is the discriminator loss?

Enter: check / next