Training pattern selection consists of selecting a subset of patterns from
a given training set so that the size of the set is reduced while its repre
sentational power is not affected, Training pattern selection is especially
desirable for the effective operation of nearest-neighbour-based decision
systems and for fast training of artificial neural networks. Counter-cluste
ring is proposed here for training pattern selection, Based on a harmony-th
eory artificial neural network, counter-clustering exposes the clusters tha
t exist within each class of the training set and provides a measure of the
interior/exterior nature of each training pattern. Training pattern select
ion is, subsequently, accomplished one shot, by retaining the boundary, iso
lated and exterior training patterns, while discarding most of the interior
training patterns from each cluster. Examples of training pattern selectio
n via counter-clustering are presented; the corresponding pattern classific
ation results are reported and evaluated.