In this paper, a new model of Probabilistic Model-Building Genetic Algorithms (PMBGAs), Distributed PMBGA (DPMBGA), is proposed. In the DPMBGA, the correlation among the design var...
Principal Component Analysis (PCA) is a basis transformation to diagonalize an estimate of the covariance matrix of input data and, the new coordinates in the Eigenvector basis ar...
A novel backwards viewpoint of Principal Component Analysis is proposed. In a wide variety of cases, that fall into the area of Object Oriented Data Analysis, this viewpoint is se...
— Over the last century, Component Analysis (CA) methods such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Canonical Correlation Analysis (CCA), Lap...
Manifolds are widely used to model non-linearity arising in a range of computer vision applications. This paper treats statistics on manifolds and the loss of accuracy occurring wh...