COST FUNCTION AND PATTERN DISTRIBUTION OF THE BAYESIAN PERCEPTRON

We determine the cost function that minimizes the generalization error of a perceptron learning a realizable task. This cost endows the perc epton with the optimal (Bayesian) generalization performance. The dist ribution of distances of the training patterns of the optimal generali zer is determined. (C) 1997 Elsevier Science B.V.