Three problems associated with uncertainty in feedforward neural network predictions are discussed. First, the uncertainty present in the input vector propagates through a trained network into the output vector, and this uncer- tainty is determined using the matrix of partial derivatives defining the change in each output with respect to the inputs. Second, because the partial derivative information conveys the relative sensitivity of a given output to each of the inputs, it can be used as a tool to determine the relevance of each of the inputs to the out prediction. Finally, the influence of random choices for training and testing data sets is investigated. The variability in these solutions provides a measure of the fossilized bias error in the network with respect to development decisions. The approaches are illustrated using examples of four-quadrant propeller predictions.
Uncertainty Analysis Applied to Feedforward Neural Networks
Ship Technology Research ; 54 , 3 ; 114-124
2007-07-01
11 pages
Article (Journal)
Electronic Resource
English
Uncertainty Analysis Applied to Feedforward Neural Networks
Online Contents | 2007
|Adaptive Type Feedforward Feedback Controller Using Neural Networks
British Library Online Contents | 1994
|Efficient Supervised Learning of Multilayer Feedforward Neural Networks
British Library Conference Proceedings | 1994
|Feedforward Control under the Presence of Uncertainty
British Library Online Contents | 2004
|