ml-recall

#minimax-value

1 question from Generative Adversarial Networks

Timed test

1 question · 1 min · score and explanations at the end

Take a test

Q 1 / 1Score 0 / 0

Shuffle questions

Question 1: Short answer

hardAI-checkedGenerative Adversarial Networks

For the original GAN, the value of the game at the optimal discriminator is $C(G) = \max_D V(D, G) = -\log 4 + 2\,\mathrm{JSD}(p_{\text{data}} \,\|\, p_g)$ (natural log). Its minimum, $-\log 4$ , is reached when $p_g = p_{\text{data}}$ . What is $C(G)$ when $p_{\text{data}}$ and $p_g$ have disjoint supports?

Enter: check / next