It would be interesting to use on-line updating for feature extraction,
its main uses would be in cases where data change over time (PCA and
autoencoders are very close to each other). It was a poor choice of
words when I used 'naturally batch', stochastic updating can definitely
be supported, but I think its much less used than algorithms such as
Stochastic Gradient Descent or Multi-Layer Perceptron, but ofcourse I
could be wrong here.
Thanks!
On 6/27/2013 6:05 AM, Vlad Niculae wrote:
Why would autoencoders be naturally batch?
------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:
Build for Windows Store.
http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
