I think if you used a random seed for weights and training data order, and reran... | Hacker News

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

		londons_explore on Feb 19, 2024 \| parent \| context \| favorite \| on: The boundary of neural network trainability is fra... I think if you used a random seed for weights and training data order, and reran the experiment enough times to average out the noise, then the resulting charts would then be smooth with no fractal patterns.

baq on Feb 19, 2024 | [–]

An interesting conjecture, well worth a paper in response.

catlifeonmars on Feb 19, 2024 | [–]

Do you consider the random seed (or by extension the randomized initial weights) a hyperparameter?

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