Neural Computation
January 1996, Vol. 8, No. 1, Pages 152-163
(doi: 10.1162/neco.1996.8.1.152)
A Comparison of Some Error Estimates for Neural Network Models
Article PDF (513.42 KB)
Abstract
We discuss a number of methods for estimating the standard error of predicted values from a multilayer perceptron. These methods include the delta method based on the Hessian, bootstrap estimators, and the “sandwich” estimator. The methods are described and compared in a number of examples. We find that the bootstrap methods perform best, partly because they capture variability due to the choice of starting weights.