Parallel computation in computer simulation for neural networks

Author

Houjin, Chen ; Baozong, Yuan ; Baxter, D.A. ; Byrne, J.H.

Author_Institution

Inst. of Inf. Sci., Northern Jiaotong Univ., Beijing, China

Volume

1

fYear

2002

fDate

28-31 Oct. 2002

Firstpage

641

Abstract

A neural network consists of many neurons connected by synapses. It is able to process information and generate a specific pattern of electrical activity. To analyze its interior structure and exterior properties, one computer simulator was implemented. As the neural network is complicated and the simulation is time consuming, so data structure and computation scheme are critical to the simulation process. In this paper, the data structure is designed to be sharable and reusable, and parallel computation is applied to match the data structure. With the data structure and parallel computation, the simulation is made much more efficient and high performance was achieved.

Keywords

data structures; digital simulation; neural nets; parallel processing; computation scheme; computer simulation; exterior properties; interior structure; neural networks; parallel computation; performance; reusable data structure; sharable data structure; Biomembranes; Computational modeling; Computer networks; Computer simulation; Concurrent computing; Data structures; Intelligent networks; Neural networks; Neurons; Voltage;

fLanguage

English

Publisher

ieee

Conference_Titel

TENCON '02. Proceedings. 2002 IEEE Region 10 Conference on Computers, Communications, Control and Power Engineering

Print_ISBN

0-7803-7490-8

Type

conf

DOI

10.1109/TENCON.2002.1181356

Filename

1181356