Visualizing a Trained Autoencoder

From Ufldl

Jump to: navigation, search
m
Line 10: Line 10:
parameters <math>\textstyle W^{(1)}_{ij}</math> (ignoring
parameters <math>\textstyle W^{(1)}_{ij}</math> (ignoring
the bias term for now) using a 2D image.  In particular, we think of
the bias term for now) using a 2D image.  In particular, we think of
-
<math>\textstyle a^{(1)}_i</math> as some non-linear feature of the input <math>\textstyle x</math>.
+
<math>\textstyle a^{(2)}_i</math> as some non-linear feature of the input <math>\textstyle x</math>.
We ask:
We ask:
What input image <math>\textstyle x</math> would cause
What input image <math>\textstyle x</math> would cause
-
<math>\textstyle a^{(1)}_i</math> to be maximally activated?
+
<math>\textstyle a^{(2)}_i</math> to be maximally activated?
(Less formally, what is the feature that hidden unit <math>\textstyle i</math> is looking for?)
(Less formally, what is the feature that hidden unit <math>\textstyle i</math> is looking for?)
For this question to have a non-trivial answer,
For this question to have a non-trivial answer,

Revision as of 00:33, 4 May 2011

Personal tools