Timeline for Why do we need sparsity for auto-encoders?

8 events

when toggle format	what		by	license	comment
S Jul 12, 2015 at 7:42	history	suggested	Frames Catherine White		added tag autoencoders
Jul 12, 2015 at 7:34	review	Suggested edits
S Jul 12, 2015 at 7:42
Jul 12, 2015 at 5:01	answer	added	Charlie Parker		timeline score: 4
Jul 12, 2015 at 4:50	comment	added	Charlie Parker		I'd guess its because otherwise the auto-encoder would learn the identity mapping. Though, I forget the details, though the intuition is very simple, if you don't put any kind of constraints on the auto-encoder, then if the auto-encoder is trying to minimize some kind of cost function $\frac{1}{n}\sum_{t} \\| x^{(t)} - \hat{x}^{(t)}\\|_{\mathcal{H}}$, the it might be that $x^{(t)} - \hat{x}^{(t)} = 0 \iff x^{(t)} = \hat{x}^{(t)}$ is the obvious/trivial/vacuous minimizer.
S Jul 21, 2012 at 21:14	history	suggested	jonsca	CC BY-SA 3.0	Grammar cleanup
Jul 21, 2012 at 21:09	review	Suggested edits
S Jul 21, 2012 at 21:14
Jul 21, 2012 at 16:34	history	tweeted			twitter.com/#!/StackStats/status/226716834311839745
Jul 21, 2012 at 15:15	history	asked	Upul	CC BY-SA 3.0