DocumentCode :
2959729
Title :
A constrained-optimization approach to training neural networks for smooth function approximation and system identification
Author :
Di Muro, Gianluca ; Ferrari, Silvia
Author_Institution :
Mech. Eng., Duke Univ., Durham, NC
fYear :
2008
fDate :
1-8 June 2008
Firstpage :
2353
Lastpage :
2359
Abstract :
A constrained-backpropagation training technique is presented to suppress interference and preserve prior knowledge in sigmoidal neural networks, while new information is learned incrementally. The technique is based on constrained optimization, and minimizes an error function subject to a set of equality constraints derived via an algebraic training approach. As a result, sigmoidal neural networks with long term procedural memory (also known as implicit knowledge) can be obtained and trained repeatedly on line, without experiencing interference. The generality and effectiveness of this approach is demonstrated through three applications, namely, function approximation, solution of differential equations, and system identification. The results show that the long term memory is maintained virtually intact, and may lead to computational savings because the implicit knowledge provides a lasting performance baseline for the neural network.
Keywords :
algebra; backpropagation; differential equations; function approximation; neural nets; optimisation; algebraic training; constrained-backpropagation training technique; constrained-optimization approach; differential equations; equality constraints; interference suppression; neural networks training; sigmoidal neural networks; smooth function approximation; system identification; Artificial neural networks; Biological neural networks; Constraint optimization; Function approximation; Interference constraints; Interference suppression; Mechanical engineering; Neural networks; Neurons; System identification;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 2008. IJCNN 2008. (IEEE World Congress on Computational Intelligence). IEEE International Joint Conference on
Conference_Location :
Hong Kong
ISSN :
1098-7576
Print_ISBN :
978-1-4244-1820-6
Electronic_ISBN :
1098-7576
Type :
conf
DOI :
10.1109/IJCNN.2008.4634124
Filename :
4634124
Link To Document :
بازگشت