A kernel Principal Component Analysis (kPCA) digest with a new backward mapping (pre-image reconstruction) strategy

Methodologies for multidimensionality reduction aim at discovering low-dimensional manifolds where data ranges. Principal Component Analysis (PCA) is very effective if data have linear structure. But fails in identifying a possible dimensionality reduction if data belong to a nonlinear low-dimensional manifold. For nonlinear dimensionality reduction, kernel Principal Component Analysis (kPCA) is appreciated because of its simplicity and ease implementation. The paper provides a concise review of PCA and kPCA main ideas, trying to collect in a single document aspects that are often dispersed. Moreover, a strategy to map back the reduced dimension into the original high dimensional space is also devised, based on the minimization of a discrepancy functional.

[1]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[2]  D. González,et al.  kPCA-Based Parametric Solutions Within the PGD Framework , 2017 .

[3]  Yasuhiro Date,et al.  Application of kernel principal component analysis and computational machine learning to exploration of metabolites strongly associated with diet , 2018, Scientific Reports.

[4]  Pedro Díez,et al.  A note on the convergence of the secant method for simple and multiple roots , 2003, Appl. Math. Lett..

[5]  Klaus-Robert Müller,et al.  Kernels, Pre-images and Optimization , 2013, Empirical Inference.

[6]  Gunnar Rätsch,et al.  Kernel PCA and De-Noising in Feature Spaces , 1998, NIPS.

[7]  Sergio Zlotnik,et al.  Algebraic PGD for tensor separation and compression: An algorithmic approach , 2018, Comptes Rendus Mécanique.

[8]  E. Lopez,et al.  A Manifold Learning Approach for Integrated Computational Materials Engineering , 2017 .

[9]  Allen Tannenbaum,et al.  Statistical shape analysis using kernel PCA , 2006, Electronic Imaging.

[10]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[11]  Gunnar Rätsch,et al.  Kernel PCA pattern reconstruction via approximate pre-images. , 1998 .

[12]  Marco Viceconti,et al.  Comprehensive evaluation of PCA-based finite element modelling of the human femur. , 2014, Medical engineering & physics.

[13]  Quan Wang,et al.  Kernel Principal Component Analysis and its Applications in Face Recognition and Active Shape Models , 2012, ArXiv.

[14]  Ivor W. Tsang,et al.  The pre-image problem in kernel methods , 2003, IEEE Transactions on Neural Networks.

[15]  P. Dez,et al.  A Note on the Convergence of the Secant Method for Simple and Multiple Roots , 2004 .

[16]  P. G. Ciarlet,et al.  Introduction to Numerical Linear Algebra and Optimisation , 1989 .

[17]  Jian-Huang Lai,et al.  Penalized Preimage Learning in Kernel Principal Component Analysis , 2010, IEEE Transactions on Neural Networks.

[18]  Sabine Van Huffel,et al.  Application of Kernel Principal Component Analysis for Single-Lead-ECG-Derived Respiration , 2012, IEEE Transactions on Biomedical Engineering.

[19]  Christopher J. Taylor,et al.  Kernel Principal Component Analysis and the construction of non-linear Active Shape Models , 2001, BMVC.

[20]  John H. Maindonald,et al.  Modern Multivariate Statistical Techniques: Regression, Classification and Manifold Learning , 2009 .