flâneur

svd

people.eecs.berkeley.edu · 1,720 words · saved by 1 readers

N/A

The Singular Value Decomposition; Clustering 127 21 The Singular Value Decomposition; Clustering THE SINGULAR VALUE DECOMPOSITION (SVD) [and its Application to PCA] Problems with PCA: Computing X > X takes ⇥(nd2 ) time. X > X is poorly conditioned ! numerically inaccurate eigenvectors. [The SVD improves both these problems.] [Earlier this semester, we talked about the eigendecomposition of a square, symmetric matrix. Unfortu- nately, nonsymmetric matrices don’t have nice…

related reading