News for hackers. Stuff that matters.

A comment on Why deep learning works even though it shouldn’t. Parent comment

Re:Why deep learning works even though it shouldn’t by ngrilly on Tuesday October 20, 2020 @03:00PM (#24841133)
> if you have enough lines you can approximate any curve within a margin

How avoid overfitting the training data?

[ Parent | on HN ]
Re:Why deep learning works even though it shouldn’t by RandyRanderson on Tuesday October 20, 2020 @09:50PM (#24844129)
You're right - a LOT of work goes into this (eg GANs) because these large nNets just start "remembering" the training data.
[ Parent | on HN ]