Fine-tuning Stacked AEs

From Ufldl

Jump to: navigation, search
(Finetuning for Classification Error)
 
Line 34: Line 34:
{{Quote|
{{Quote|
-
Note: While one could consider the softmax classifier as an additional layer, the derivation above does not. Specifically, we consider the "last layer" of the network to be the features that goes into the softmax classifier. Therefore, the derivatives (in Step 2) are computed using <math>\delta^{(n_l)} = - (\nabla_{a^{n_l}}J) \bullet f'(z^{(n_l)})</math>.
+
Note: While one could consider the softmax classifier as an additional layer, the derivation above does not. Specifically, we consider the "last layer" of the network to be the features that goes into the softmax classifier. Therefore, the derivatives (in Step 2) are computed using <math>\delta^{(n_l)} = - (\nabla_{a^{n_l}}J) \bullet f'(z^{(n_l)})</math>, where  <math>\nabla J = \theta^T(I-P)</math>.
}}
}}
 +
 +
 +
{{CNN}}
 +
 +
 +
{{Languages|微调多层自编码算法|中文}}

Latest revision as of 04:04, 8 April 2013

Personal tools