Analysis on Influence of CMAC Neural Network Parameters Selection on Network Performance

Author

He, Lian-yun

Author_Institution

Dept. of Mech. & Electron. Eng., Dezhou Univ., Dezhou, China

Volume

3

fYear

2009

fDate

14-16 Aug. 2009

Firstpage

490

Lastpage

494

Abstract

CMAC neural network is a neural network to realize associative memory through various mappings, which can realize on-line learning with fast learning speed, high precision and fast convergence speed and has unique advantage in the aspect of space mapping compared with the other networks. In order to study the influence of parameters selection on the network mapping performance, the influence of CMAC network parameters on the network learning speed, convergence speed and system error is analyzed by taking the control of double inverted pendulum as an example.

Keywords

cerebellar model arithmetic computers; computer network performance evaluation; convergence; learning (artificial intelligence); network parameters; parameter estimation; CMAC neural network; associative memory; cerebellar model articulation controller; convergence speed; double inverted pendulum control; network learning speed; network mapping performance; on-line learning; parameters selection; space mapping; system error; Artificial neural networks; Associative memory; Brain modeling; Computer networks; Convergence; Helium; High performance computing; Neural networks; Performance analysis; Quantization; CMAC neural network; Convergence speed; Learning speed; Network parameters; System error;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation, 2009. ICNC '09. Fifth International Conference on

Conference_Location

Tianjin

Print_ISBN

978-0-7695-3736-8

Type

conf

DOI

10.1109/ICNC.2009.462

Filename

5365567