The PSOM Algorithm and Applications

Abstract: In this paper we discuss the “Parameterized Selforganizing Maps” (PSOM) as a learning method for rapidly creating high-dimensional, continuous mappings. The PSOM can be viewed as the continous generalization of the discrete topology preserving map build by Kohonen’s SOM algorithm. By making use of available topological information the PSOM shows excellent generalization capabilities from a small set of training data. Unlike most other existing approaches that are limited to the representation of input-output mappings, the PSOM provides as an important generalization a flexibly usable, continuous associate memory. This allows to represent several related mappings – coexisting in a single and coherent framework. We present application examples for simultaneous learning of robot forward and backward kinematics, concepts for an integrated redundancy control scheme, and the application to gesture recognition in a humanmachine-interface. All profit from the approximation accuracy gained from only a few training examples.