Numerical and qualitative stability analysis of ring and linear neural networks with a large number of neurons (Q2912235)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 6082517
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Numerical and qualitative stability analysis of ring and linear neural networks with a large number of neurons |
scientific article; zbMATH DE number 6082517 |
Statements
14 September 2012
0 references
neural networks
0 references
delay differential matrix equations
0 references
stability cone
0 references
delay-independent stability
0 references
Numerical and qualitative stability analysis of ring and linear neural networks with a large number of neurons (English)
0 references
This paper considers a large number of ``neurons'' arranged in either a ring or line, each coupled to both of its nearest neighbours, but possibly in a delayed fashion. The dynamics are linear, so there is only one fixed point, the origin. Its stability depends on the strength of coupling between neighbouring neurons (which may be different for neurons to the right as opposed to to the left) and the values of any delays. Some theorems regarding the dependence of stability on these parameter values are proven, and numerical results given. Some of the results are given in the limit as the number of neurons tends to infinity.
0 references
0.7917985320091248
0 references
0.7793985605239868
0 references
