Why don't we use non-constant learning rates for gradient decent for things other then neural networks? – stats.stackexchange.com

Why don't we use non-constant learning rates for gradient decent for things other then neural networks? – stats.stackexchange.com

11:14 No Comments
Deep learning literature is full of clever tricks with using non-constant learning rates in gradient descent. Things like exponential decay, RMSprop, Adagrad etc. are easy to implement and are ...

from Hot Questions - Stack Exchange OnStackOverflow
via Blogspot

Share this

Unknown

Artikel Terkait

0 Comment to "Why don't we use non-constant learning rates for gradient decent for things other then neural networks? – stats.stackexchange.com"

Post a Comment

Subscribe to: Post Comments (Atom)