Nonlinear Component Analysis as a Kernel Eigenvalue Problem
Bernhard Schölkopf, Alexander Johannes Smola, Klaus‐Robert Müller · Neural Computation · 1998
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.
Cited by 8,084 works. doi:10.1162/089976698300017467
Authors
References
- The Nature of Statistical Learning Theory
- Support-Vector Networks
- Backpropagation Applied to Handwritten Zip Code Recognition
- Methods of Mathematical Physics
- A training algorithm for optimal margin classifiers
- Principal Component Analysis
- Theoretical Foundations of the Potential Function Method in Pattern Recognition Learning
- Application of the Karhunen-Loeve procedure for the characterization of human faces
- Statistical pattern recognition: a review
- Automation and Remote Control