Quantcast
Channel: Machine Learning
Viewing all articles
Browse latest Browse all 62787

Why use sparse overcomplete autoencoders vs. non-over complete normal ones?

$
0
0

This page recommends the use of sparsified overcomplete autoencoders (i.e. the learned representation has a higher dimensionality than the input, but many of the values are zeros).

What is the advantage of this over a normal autoencoder?

submitted by jamesmcm
[link][4 comments]

Viewing all articles
Browse latest Browse all 62787

Trending Articles