Kernel Principal Component Analysis

A new method for performing a nonlinear form of Principal Component Analysis is proposed. By the use of integral operator kernel functions, one can e ciently compute principal components in high dimensional feature spaces, related to input space by some nonlinear map; for instance the space of all possible d pixel products in images. We give the derivation of the method and present experimental results on polynomial feature extraction for pattern recognition.

• Type: Conference
• Year: 1997
• Where: ICANN
• Authors: Bernhard Schölkopf, Alex J. Smola, Klaus-Robert Müller
• Keywords: Compute Principal Components, ICANN 1997, Neural Networks, Principal Component Analysis, Principal Components