Exercise:Independent Component Analysis
From Ufldl
(→Step 4: Optimization) |
|||
Line 47: | Line 47: | ||
[[File:OrthonormalICAFeatures.png | 350px]] | [[File:OrthonormalICAFeatures.png | 350px]] | ||
- | + | It is comparatively difficult to optimize for the objective while enforcing the orthonormality constraint using gradient descent, and convergence can be slow. Hence, in situations where an orthonormal basis is not required, other faster methods of learning bases (such as [[Sparse Coding: Autoencoder Interpretation | sparse coding]]) may be preferable. | |
=== Appendix === | === Appendix === |