ml-recall

#l1-loss

2 questions from Generative Adversarial Networks, PyTorch

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 translates a real image $x$ to the other domain and back, giving $\hat{x}$ . Its cycle-consistency loss is $\lambda$ times the mean absolute pixel difference, with $\lambda = 10$ . For a toy 4-pixel image $x = (0.5, -0.2, 0.1, 0.0)$ and reconstruction $\hat{x} = (0.4, -0.2, 0.3, 0.1)$ , what is the loss?

Enter: check / next