Title :
Automated stopping criteria for neural network training
Author :
Natarajan, Siva ; Rhinehart, R. Russell
Author_Institution :
Dept. of Chem. Eng., Texas Tech. Univ., Lubbock, TX, USA
Abstract :
A novel technique for improving neural network (NN) training is developed and demonstrated. This technique makes more efficient use of the available data for training and also automates the decision criteria to stop training
Keywords :
learning (artificial intelligence); neural nets; automated stopping criteria; neural network training; Automatic control; Chemical engineering; Costs; Fault diagnosis; Feedforward neural networks; Least squares methods; Neural networks; Process control; Recurrent neural networks; System identification;
Conference_Titel :
American Control Conference, 1997. Proceedings of the 1997
Conference_Location :
Albuquerque, NM
Print_ISBN :
0-7803-3832-4
DOI :
10.1109/ACC.1997.609154
