ml-recall

#discriminator-loss

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

mediumAI-checkedGenerative Adversarial Networks

d2l trains the discriminator on a real and a fake batch with the binary cross-entropy averaged over the two, $\frac{1}{2}\left[-\log D(\mathbf{x}) - \log\left(1 - D(G(\mathbf{z}))\right)\right]$ per pair of examples. At the equilibrium where the discriminator outputs 0.5 for every input, what is this loss (natural logarithm)?

Enter: check / next