Title :
On the Optimum Method of Feedforward Multi-Layer Neural Network
Author :
Ying, Hu ; Jin, Huang
Author_Institution :
Dalian Maritime Univ., Dalian
Abstract :
Defining network average error as optimum objective function, weights and thresholds as design variable, which are rationally sorted, a new kind of real conjugate terraced optimum algorithm is studied. Compared with BP algorithm, the compute time is reduced and the precision is improved. A computing program about weights and threshold, based on high precision conjugate gradient optimum algorithm of multi-layer neural network, is put forward and programmed. The selecting method of rational construct is also pointed out. Through an application instance, its advantage and applying prospect is validated.
Keywords :
conjugate gradient methods; multilayer perceptrons; conjugate gradient optimum algorithm; feedforward multilayer neural network; network average error; optimum objective function; real conjugate terraced optimum algorithm; Algorithm design and analysis; Computer networks; Convergence; Electronic mail; Error correction; Multi-layer neural network; Neural networks; Optimization methods; Pattern recognition; Surges; Conjugate gradient optimum algorithm; Multi-layer NN; Rational construct optimizations;
Conference_Titel :
Control Conference, 2007. CCC 2007. Chinese
Conference_Location :
Hunan
Print_ISBN :
978-7-81124-055-9
Electronic_ISBN :
978-7-900719-22-5
DOI :
10.1109/CHICC.2006.4346849
