An Expectation-Maximization Approach to Nonlinear Component Analysis

Authors:
Roman Rosipal;Mark Girolami
Affiliations:
Computational Intelligence Research Unit, Department of Computing and Information Systems, University of Paisley, Paisley, PA1 2BE, Scotland, U.K.;Computational Intelligence Research Unit, Department of Computing and Information Systems, University of Paisley, Paisley, PA1 2BE, Scotland, U.K.
Venue:
Neural Computation
Year:
2001

Citing 3
Cited 15

Nonlinear component analysis as a kernel eigenvalue problem

Neural Computation
Advances in kernel methods: support vector learning

Advances in kernel methods: support vector learning
A unifying review of linear Gaussian models

Neural Computation

Orthogonal series density estimation and the kernel eigenvalue problem

Neural Computation
Bayesian framework for least-squares support vector machine classifiers, Gaussian processes, and kernel fisher discriminant analysis

Neural Computation
A constrained EM algorithm for principal component analysis

Neural Computation
An Improved Algorithm for Kernel Principal Component Analysis

Neural Processing Letters
Letters: A method for speeding up feature extraction based on KPCA

Neurocomputing
Efficiently updating and tracking the dominant kernel principal components

Neural Networks
Letters: An unified EM algorithm for PCA and KPCA

Neurocomputing
Efficient tracking of the dominant eigenspace of a normalized kernel matrix

Neural Computation
Nonlinear Component Analysis for Large-Scale Data Set Using Fixed-Point Algorithm

ISNN 2009 Proceedings of the 6th International Symposium on Neural Networks: Advances in Neural Networks - Part III
Matrix-based kernel principal component analysis for large-scale data set

IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
Edge detection in the feature space

Image and Vision Computing
A Fast Algorithm for Updating and Downsizing the Dominant Kernel Principal Components

SIAM Journal on Matrix Analysis and Applications
Kernel principal component analysis for large scale data set

ICIC'06 Proceedings of the 2006 international conference on Intelligent Computing - Volume Part I
A fast feature extraction method for kernel 2DPCA

ICIC'06 Proceedings of the 2006 international conference on Intelligent Computing - Volume Part I
Adaptive kernel subspace method for speeding up feature extraction

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The proposal of considering nonlinear principal component analysis as a kernel eigenvalue problem has provided an extremely powerful method of extracting nonlinear features for a number of classification and regression applications. Whereas the utilization of Mercer kernels makes the problem of computing principal components in, possibly, infinite-dimensional feature spaces tractable, there are still the attendant numerical problems of diagonalizing large matrices. In this contribution, we propose an expectation-maximization approach for performing kernel principal component analysis and show this to be a computationally efficient method, especially when the number of data points is large.