Training Neural Networks with Deficient Data

Part of Advances in Neural Information Processing Systems 6 (NIPS 1993)

Authors

Volker Tresp, Subutai Ahmad, Ralph Neuneier

Abstract

We analyze how data with uncertain or missing input features can be incorporated into the training of a neural network. The gen(cid:173) eral solution requires a weighted integration over the unknown or uncertain input although computationally cheaper closed-form so(cid:173) lutions can be found for certain Gaussian Basis Function (GBF) networks. We also discuss cases in which heuristical solutions such as substituting the mean of an unknown input can be harmful.