DocumentCode
3231969
Title
Evolving a strongly recurrent neural network to simulate biological neurons
Author
Soule, Terence ; Chen, YingYin ; Wells, Richard B.
Author_Institution
Idaho Univ., Moscow, ID, USA
Volume
4
fYear
2002
fDate
5-8 Nov. 2002
Firstpage
3191
Abstract
In this research we use evolutionary techniques to evolve recurrent neural networks that produce a pulsed output when triggered by a constant valued input. Networks of several different sizes and configurations are successfully evolved demonstrating that this is a robust technique. The resultant networks can be used as approximations of certain types of biological neurons or of central pattern generators.
Keywords
genetic algorithms; recurrent neural nets; biological neurons simulation; biomimetic neurons; central pattern generators; constant valued input; evolutionary computation; genetic algorithms; pulsed output; recurrent neural network; robust technique; Biological information theory; Biological system modeling; Computational modeling; Computer science; Evolutionary computation; Frequency; Legged locomotion; Motion control; Neurons; Recurrent neural networks;
fLanguage
English
Publisher
ieee
Conference_Titel
IECON 02 [Industrial Electronics Society, IEEE 2002 28th Annual Conference of the]
Print_ISBN
0-7803-7474-6
Type
conf
DOI
10.1109/IECON.2002.1182908
Filename
1182908
Link To Document