Parallel training for neural networks using PVM with shared memory

Author

Araiijo, M.A.A. ; Teixeira, Edilberto P. ; Camargo, Fábio R. ; Almeida, João P V

Author_Institution

UFU, Goiania, GO, Brazil

Volume

2

fYear

2003

fDate

8-12 Dec. 2003

Firstpage

1315

Abstract

We present a peculiar parallel implementation of artificial neural networks using the backpropagation training algorithm. The message pass interface PVM is used in the Linux operating system environment, implemented in a cluster of IBM-PC machines. An optimized object-oriented framework to train neural networks, developed in C++, is part of the system presented. A shared memory framework was implemented to improve the training phase. One of the advantages of the system is the low cost, considering that its performance can be compared to similar powerful parallel machines.

Keywords

C++ language; backpropagation; message passing; neural nets; operating systems (computers); parallel machines; shared memory systems; IBM-PC machines; Linux operating system environment; PVM framework; artificial neural networks; backpropagation training algorithm; message pass interface; optimized object-oriented framework; parallel computing; parallel machines; shared memory framework; Application software; Artificial neural networks; Backpropagation algorithms; Computer networks; Costs; Industrial training; Linux; Neural networks; Neurons; Operating systems;

fLanguage

English

Publisher

ieee

Conference_Titel

Evolutionary Computation, 2003. CEC '03. The 2003 Congress on

Print_ISBN

0-7803-7804-0

Type

conf

DOI

10.1109/CEC.2003.1299821

Filename

1299821