A neural network approach to adaptive pattern analysis - the deformable feature map

Abstract. In this paper, we presen t an algorithm that provides adaptive plasticity in function approximation problems: the deformable (feature) map (DM) algorithm. The DM approach reduces a class of similar function approximation problems to the explicit supervised one-shot training of a single data set. This is followed by asubsequent, appropriate similarity transformation whic his based on a self-organized deformation of the underlying multidimensional probability distributions. After discussing the theory of the DM algorithm, w euse a computer sim ulation to visualize its e ects on a tw o-dimensional toy example. Finally, we presen t results of its application to the real-world problem of fully automatic voxel-based multispectral image segmentation, employing magnetic resonance data sets of the human brain.

• Type: Conference
• Year: 2000
• Where: ESANN
• Authors: Axel Wismüller, Frank Vietze, Dominik R. Dersch, Klaus Hahn, Helge Ritter
• Keywords: DM Algorithm, ESANN 2000, ESANN 2007, Function Approximation Problems, Similar Function Approximation