Abstract
In this paper we discuss training of three-layered neural network classifiers by solving inequalities. Namely, first we represent each class by the center of the training data belonging to the class, and determine the set of hyperplanes that separate each class (i.e., each center) into a single region. Then according to whether the center is on the positive or negative side of the hyperplane, we determine the target values of each class for the hidden neurons (i.e., hyperplanes). Since the convergence condition of the neural network classifier is now represented by the two sets of inequalities, we solve the sets successively by the Ho-Kashyap algorithm. We demonstrate the advantage of our method over the backpropagation algorithm using several benchmark data sets.
