Generating arbitrary rhythmic patterns with purely inhibitory neural networks

Yang, Zhijun and Franca, Felipe (1998) Generating arbitrary rhythmic patterns with purely inhibitory neural networks. In: ESANN'1998 proceedings - 6th European symposium on artificial neural networks (Bruges, 22-23-24 April 1998). Verleysen, Michel, ed. D facto, pp. 53-58. ISBN 9782960004984

Full text is not in this repository.

Abstract

A novel approach for the prediction and generation of coupled neural oscillation among arbitrarily connected inhibitory neurons is proposed. Based on Scheduling by Multiple Edge Reversal (SMER), a very simple distributed algorithm, neural network building blocks can be configured for the generation of complex rhythmic patterns with a very high independence from individual neuronal models. A method for the organization and simulation of the new approach is illustrated by mimicking the main rhythmic gait patterns of an hexapodal animal.

Item Type: Book Section
Research Areas: A. > School of Science and Technology > Design Engineering and Mathematics
Item ID: 9929
Useful Links:
Depositing User: Zhijun Yang
Date Deposited: 20 Feb 2013 13:24
Last Modified: 13 Oct 2016 14:25
URI: https://eprints.mdx.ac.uk/id/eprint/9929

Actions (login required)

Edit Item Edit Item