Hybrid hardware for a highly parallel search in the context of learning classifiers

Based on a comparison of input data with a set of prototypes, classifier sy stems identify the most appropriate representative for a given sample patte rn. One remarkable classifier is Kohonen's Self-Organizing Map and the rela ted learning vector quantizer, as these algorithms are highly parallel. For real-time applications the classifier search may be one of the time critic al processes. We discuss specialized hardware being able to execute such a search in a fully parallel manner. Also the learning and updating of protot ypes is performed in parallel controlled by a propagating front. Finally, w e present experimental results concerning an unsupervised learning vector q uantizer (LVQ) and a self-organizing map (SOM) obtained from our thyristor- based analog-digital hybrid system. (C) 2001 Elsevier Science B.V. All righ ts reserved.