Improved Feature Extraction Using Weightless Neural Networks(IWNC)

Abstract

The weightless neural classifier (WNC) is based on the collective response of RAM-based neurons. The ability of producing prototypes, analog to unconstrained images, from learned categories, was first introduced in the (IWNC) model. By counting the frequency of write accesses at each RAM neuron during the training phase, it is possible to associate the most accessed addresses to the corresponding input field contents that defined them. This work is about extracting information from such frequency counting in the form of fuzzy rules as an alternative way to describe the same images produced by (IWNC) as logical prototypes.