FractionalRadix commented on Neural Networks: Algorithms, Applications, and Programming Techniques by James A. Freeman (Computation and Neural Systems Series)
Chapter 6 discusses the Counter Propagation Network (CPN). Although not stated explicitly, it is suggested that it mainly classifies vectors. The "instar" and "outstar" constructions are introduced. It is pointed out that there are some problems a CPN can't learn that a BPN (Back Propagation Network) can: counting the number of 1's in a binary vector, because vectors that differ in only 1 bit should be put in different classes.