Title :
Minimizing risk using prediction uncertainty in neural network estimation fusion and its application to papermaking
Author :
Edwards, Peter J. ; Peacock, Andrew M. ; Renshaw, David ; Hannah, John M. ; Murray, Alan F.
Author_Institution :
Dept. of Electron. & Electr. Eng., Edinburgh Univ., UK
fDate :
5/1/2002 12:00:00 AM
Abstract :
The paper presents Bayesian information fusion theory in the context of neural-network model combination. It shows how confidence measures can be combined with individual model estimates to minimize risk through the fusion process. The theory is illustrated through application to the real task of quality prediction in the papermaking industry. Prediction uncertainty estimates are calculated using approximate Bayesian learning. These are incorporated into model combination as confidence measures. Cost functions in the fusion center are used to control the influence of the confidence measures and improve the performance of the resultant committee
Keywords :
Bayes methods; manufacturing data processing; neural nets; paper industry; risk management; uncertainty handling; Bayesian information fusion theory; Bayesian learning; approximate Bayesian learning; confidence measures; cost functions; fusion center; fusion process; individual model estimates; model combination; neural network estimation fusion; neural network model combination; papermaking industry; prediction uncertainty; prediction uncertainty estimates; prediction uncertainty estimation; quality prediction; risk minimization; Bayesian methods; Context modeling; Cost function; Industrial training; Intelligent networks; Neural networks; Predictive models; Pulp and paper industry; Risk management; Uncertainty;
Journal_Title :
Neural Networks, IEEE Transactions on
DOI :
10.1109/TNN.2002.1000137
