Etienne Barnard and Elizabeth C. Botha,
Back-Propagation Uses Prior Information Efficiently,
IEEE Transactions on Neural Networks,
4, 5, pp. 794-802, September 1993.
Etienne Barnard and Elizabeth C. Botha,
Back-Propagation Uses Prior Information Efficiently,
IEEE Transactions on Neural Networks,
4, 5, pp. 794-802, September 1993.The ability of neural net classifiers to deal with a priori probabilities, and their performance on separate test sets is evaluated. It is found that back-propagation employs a priori information in a slightly suboptimal fashion, but that this does not have serious consequences on the performance of the classifier. Furthermore, it is found that the inferior generalization that results when an excessive number of network parameters are used can (partially) be ascribed to this suboptimality.