Pattern Classi cation by Duda et al. - Tommy Odland

← Back to document page

The computation is often performed using the Singular Value Decomposition (SVD). Discriminant Analysis (DA) projects to a lower dimensional subspace with optimal ... The pseudoinverse should never be used explicitly because it’s numerically wasteful and unstable. It represents the analytical solution to the problem min x

  Value, Singular, Decomposition, Singular value decomposition, Dudas, Pseudoinverse