Sequential RBF function estimator: memory regression network

Author

Chow, Chi-Kin ; Tsui, Hung-Tat

Author_Institution

Dept. of Electron. Eng., Chinese Univ. of Hong Kong

Volume

5

fYear

0

fDate

0-0 0

Firstpage

4815

Abstract

The neural-network training algorithm can be divided into 2 categories: (1) batch mode and (2) sequential mode. In this paper, a novel online RBF network called "memory regression network (MRN)" is proposed. Different from the previous approaches (de Freitas, N, et al., Aug. 1999), (Schiffman, W, et al., 1993), MRN involves two types of memories: experience and neuron, which handle short and long term memories respectively. By simulating human\´s learning behavior, a given function can be estimated without memorizing the whole training set. Two sets of function estimation experiments are examined in order to illustrate the performance of the proposed algorithm. The results show that MRN can effectively approximate the given function within a reasonable time and acceptable mean square error

Keywords

learning (artificial intelligence); mean square error methods; radial basis function networks; sequential estimation; batch mode; mean square error method; memory regression network; neural network training algorithm; sequential RBF function estimation; sequential mode; Humans; Image processing; Interpolation; Laboratories; Neural networks; Neurons; Radial basis function networks; Radio access networks; Robots; Signal processing algorithms;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Man and Cybernetics, 2004 IEEE International Conference on

Conference_Location

The Hague

ISSN

1062-922X

Print_ISBN

0-7803-8566-7

Type

conf

DOI

10.1109/ICSMC.2004.1401293

Filename

1401293