Nonlinear Component Analysis as a Kernel Eigenvalue Problem

Citations Over TimeTop 1% of 1998 papers

Abstract

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.

Related Papers

• → Nonlinear Projection Trick in Kernel Methods: An Alternative to the Kernel Trick(2013)56 cited
• → Nonlinear process monitoring using improved kernel principal component analysis(2016)3 cited
• Improving Performance of Kernel Principal Component Analysis Using Combination Kernel Functions(2004)
• → Kernel PCA Part 1: Polynomials without the Kernels(2015)1 cited
• Based on kernel principal component analysis combined kernel function algorithm(2012)