Competitive Probabilistic Principal Component Analysis Neural Networks

E. López-Rubio and J.M. Ortiz-de-Lazcano-Lobato (Spain)


Soft computing, neural networks, Principal Components Analysis (PCA), competitive learning, multispectral imaging.


We present a new neural model, which extends the classical competitive learning (CL) by performing a Probabilistic Principal Components Analysis (PPCA) at each neuron. This model represents an improvement with respect to known local PCA methods, because it is not needed to present the entire data set to the network on each computing step. This allows a fast execution, while retaining the dimensionality reduction properties of the PCA. Experimental results are presented, which show the performance of the model when we use it with multispectral image data.

Important Links:

Go Back