A Comprehensive Understanding for Radial Basis Probabilistic Neural Networks

摘要

This paper makes a profound analysis on radial basis probabilistic neural networks (RBPNN) from the viewpoint of linear algebra. Specifically, the transformation properties and internal representations of the RBPNN's are investigated in alliance with the properties of the input samples so that ones can understand and grasp the mechanisms for pattern classification and function approximation of the RBPNN's. In addition, we make an analysis on the convergent behaviours of the output class weight vectors of the RBPNN's, which can be showed to be orthogonal as well. Finally, one example for classifying five kinds of different distribution patterns are given to further support our understandings and claims.