Re:Deep learning
by Houshalter on Sunday July 26, 2015 @09:15PM
(#9952885)
I don't really understand it but I suspect there is some trick going on somewhere. I don't see how you can avoid doing at least one computation for each weight pixel pair in each convolution.
[ Parent
| on HN ]
Does this answer your question?
Re:Deep learning
by Houshalter on Wednesday July 29, 2015 @03:04AM
(#9966791)
I read it, I just don't understand it.