Improved Convergence Speed of Fully Symmetric Learning Rules for Principal Component Analysis

Ralf Möller

Fully symmetric learning rules for principal component analysis can be derived from a novel objective function suggested in our previous work. We observed that these learning rules…