Sparse Coding: Autoencoder Interpretation
From Ufldl
Line 47: | Line 47: | ||
(where <math>\sqrt{s^2 + \epsilon}</math> is shorthand for <math>\sum_k{\sqrt{s_k^2 + \epsilon}}</math>) | (where <math>\sqrt{s^2 + \epsilon}</math> is shorthand for <math>\sum_k{\sqrt{s_k^2 + \epsilon}}</math>) | ||
- | Optimizing for this objective function (using the iterative method of optimizing for <math>A</math>, then <math>s</ | + | Optimizing for this objective function (using the iterative method of optimizing for <math>A</math>, then <math>s</math>, alternately) will yield features (the basis vectors of <math>A</math>) similar to those learned using the sparse autoencoder. |
== Topographic sparse coding == | == Topographic sparse coding == |