Concretely if I use the first k eigenvectors find by PCA with a point set A,to project another sparse vector b to k dimension subspace, then use L1magic to recover b. Will this be better than a Random Projecton. L1magic is a software tool for compress sensing. b is not a vector from A, but possibly have the same distribution of A.
Take the 2minute tour
×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.
