Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
10332136 | Information Processing Letters | 2005 | 8 Pages |
Abstract
A type of spiking neural net in which the internal state functions of the neurons are represented symbolically rather than numerically is described. This makes possible subtle forms of short term memory, but also raises both computational and mathematical problems. It is shown how interval arithmetic can be used for event driven simulation. Computational times are given for some tests of an implementation resulting from these ideas. Applications are briefly discussed.
Related Topics
Physical Sciences and Engineering
Computer Science
Computational Theory and Mathematics
Authors
Carl O'Dwyer, Daniel Richardson,