This is probably an uninformed question, but why are you comparing layers' outpu... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		ProjectArcturis 3 months ago \| parent \| context \| favorite \| on: Show HN: I invented a new generative model and got... This is probably an uninformed question, but why are you comparing layers' output to ground truth? Isn't the point that ground truth is unknown?

diyer22 3 months ago [–]

During neural network training, the ground truth (GT) must be known to compute the loss.

In DDN, the GT is only used to calculate the loss and guide sampling; it never becomes an input to the model.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact