This paper introduces quantum neural networks (QNN's), a class of feed
forward neural networks (FFNN's) inherently capable of estimating the
structure of a feature space in the form of fuzzy sets. The hidden uni
ts of these networks develop quantized representations of the sample i
nformation provided by the training data set in various graded levels
of certainty. Unlike other approaches attempting to merge fuzzy logic
and neural networks, QNN's can be used in pattern classification probl
ems without any restricting assumptions such as the availability of a
priori knowledge or desired membership profile, convexity of classes,
a limited number of classes, etc. Experimental results presented here
show that QNN's are capable of recognizing structures in data, a prope
rty that conventional FFNN's with sigmoidal hidden units lack.