TY - GEN

T1 - On neural network training algorithm based on the unscented Kalman filter

AU - Li, Hongli

AU - Wang, Jiang

AU - Che, Yanqiu

AU - Wang, Haiyang

AU - Chen, Yingyuan

PY - 2010/12/22

Y1 - 2010/12/22

N2 - Neural network has been widely used for nonlinear mapping, time-series estimation and classification. The backprop-agation algorithm is a landmark of network weights training. Although the vast weights update algorithms have been developed, they are often plagued by convergence to poor local optima and low learn velocity. The unscented Kalman filter is a nonlinear parameter estimation algorithm. By means of it, weights update can be realized. Higher training velocity and mapping accuracy of network can be obtained. The numerical simulation results show the effectiveness of the algorithm compared with the standard backpropagation.

AB - Neural network has been widely used for nonlinear mapping, time-series estimation and classification. The backprop-agation algorithm is a landmark of network weights training. Although the vast weights update algorithms have been developed, they are often plagued by convergence to poor local optima and low learn velocity. The unscented Kalman filter is a nonlinear parameter estimation algorithm. By means of it, weights update can be realized. Higher training velocity and mapping accuracy of network can be obtained. The numerical simulation results show the effectiveness of the algorithm compared with the standard backpropagation.

UR - http://www.scopus.com/inward/record.url?scp=78650254218&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650254218&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:78650254218

SN - 9787894631046

T3 - Proceedings of the 29th Chinese Control Conference, CCC'10

SP - 1447

EP - 1450

BT - Proceedings of the 29th Chinese Control Conference, CCC'10

T2 - 29th Chinese Control Conference, CCC'10

Y2 - 29 July 2010 through 31 July 2010

ER -