4 ms·
I've always understood PCA as SVD on a whitened matrix. Is this too simplistic of a view to take wrt implementation? https://en.m.wikipedia.org/wiki/Whitening_
by kiernanmcgowan 9y ago
I've always understood PCA as SVD on a whitened matrix. Is this too simplistic of a view to take wrt implementation?
https://en.m.wikipedia.org/wiki/Whitening_transformation https://en.m.wikipedia.org/wiki/Whitening_transformation
- celerity 9y agoI actually touch on the relation to whitening toward the bottom of the article. You can whiten your dataset from the left singular matrix U which is directly related to PCs. Thanks for reading!