Exploring the effects of Lamarckian and Baldwinian learning in evolving recurrent neural networks

Author

Ku, Kim W C ; Mak, M.W.

Author_Institution

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

fYear

1997

fDate

13-16 Apr 1997

Firstpage

617

Lastpage

621

Abstract

A drawback of using genetic algorithms (GAs) to train recurrent neural networks is that it takes a large number of generations to evolve the networks into an optimal solution. In order to reduce the number of generations taken, the Lamarckian learning mechanism and the Baldwinian learning mechanism are embedded into a cellular GA. This paper investigates the effects of these two learning mechanisms on the convergence performance of the cellular GA. The criteria that make learning useful to GAs are also discussed. The results show that the Lamarckian mechanism is able to assist the cellular GA, while the Baldwinian mechanism fails to do so. In addition to reducing the number of generations taken, we have found that it is also possible to reduce the time taken by embedding learning into the cellular GA in an appropriate manner

Keywords

cellular automata; convergence; genetic algorithms; learning (artificial intelligence); optimisation; recurrent neural nets; Baldwinian learning; Lamarckian learning; backpropagation; cellular genetic algorithm; convergence performance; evolving recurrent neural networks; genetic algorithms; neural network training; optimal solution; time; Backpropagation algorithms; Biological cells; Cellular networks; Genetic algorithms; Genetic mutations; Intelligent networks; Learning systems; Neural networks; Recurrent neural networks; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Evolutionary Computation, 1997., IEEE International Conference on

Conference_Location

Indianapolis, IN

Print_ISBN

0-7803-3949-5

Type

conf

DOI

10.1109/ICEC.1997.592386

Filename

592386