Bernhard Schölkopf1, Alexander J. Smola2, Klaus‐Robert Müller2
1Max-Planck-Institut für biologische Kybernetik, 72076 Tübingen, Germany
2GMD First (Forschungszentrum Informationstechnik), 12489 Berlin, Germany
Tóm tắt
A new method for performing a nonlinear form of principal component analysis is proposed. By the use of integral operator kernel functions, one can efficiently compute principal components in high-dimensional feature spaces, related to input space by some nonlinear map—for instance, the space of all possible five-pixel products in 16 × 16 images. We give the derivation of the method and present experimental results on polynomial feature extraction for pattern recognition.
Từ khóa
Tài liệu tham khảo
Aizerman M., 1964, Automation and Remote Control, 25, 821
