Least Mean Square Learning in Associative Memory Networks

Brown, M. and Harris, C.J. (1992) Least Mean Square Learning in Associative Memory Networks. Int. Symp. on Intelligent Control , 531-536.


Full text not available from this repository.


This paper proposes and investigates theoretically the use of a class of neural networks called Associative Memory Networks for on-line adaptive nonlinear modelling and control. This class of networks is defined to include such algorithms as the CMAC, B-splines and Fuzzy Logic. Firstly, the algorithms are defined within a unifying framework which provides a natural decomposition for a parallel implementation. Then the modelling capabilities of the networks are investigated, and some new results about the CMAC are presented. Next the instantaneous learning rules are derived and investigated from a geometrical perspective, which allows the rate of convergence to be analysed. Also a measure of the learning interference for different set (basis function) shapes is obtained. Finally, an example of an Associative Memory Network guiding an autonomous vehicle into a slot is given.

Item Type: Conference or Workshop Item (UNSPECIFIED)
Additional Information: Organisation: IEEE Address: Glasgow, UK
Divisions : Faculty of Physical Sciences and Engineering > Electronics and Computer Science > Southampton Wireless Group
ePrint ID: 250249
Accepted Date and Publication Date:
Date Deposited: 04 May 1999
Last Modified: 31 Mar 2016 13:50
Further Information:Google Scholar
URI: http://eprints.soton.ac.uk/id/eprint/250249

Actions (login required)

View Item View Item