Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/31688
Title: A study of the leaky-integrator recurrent neural dynamics and its applications
Authors: Li, LK
Shao, S
Keywords: The leaky integrator
Recurrent neural networks
Positive invariant
Attractive set
Absolutely stable
Issue Date: 2006
Publisher: Watam Press
Source: Dynamics of continuous, discrete & impulsive systems. Series A, Mathematical analysis, 2006, v. 13, p. 353-366 How to cite?
Journal: Dynamics of continuous, discrete & impulsive systems. Series A, Mathematical analysis 
Abstract: We study the characteristics of the leaky-integrator recurrent neural network dynamics and its applications. Our results show that the set of solutions of the dynamical system is positive invariant and attractive for the continuous-time recurrent neural network model. For the discrete-time recurrent neural network model, the stability analysis has been provided. Examples are given to demonstrate how our approaches can be applied to compress the data and perform the global optimization techniques to the nonlinear regression models effectively. The method offers an ideal setting to carry out the recurrent neural network approach to different areas including engineering, business and statistics.
Description: International Workshop on Differential Equations and Dynamical Systems, Guelph, Canada, 29-31 July 2005
URI: http://hdl.handle.net/10397/31688
ISSN: 1201-3390
Appears in Collections:Journal/Magazine Article

Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

38
Last Week
0
Last month
Checked on Oct 15, 2017

Google ScholarTM

Check



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.