Complex dynamics and the structure of small neural networks

Frank Pasemann

Contact the author: Please use for correspondence this email.
Submission date: 12. Jul. 2001
Pages: 31
published in: Network : computation in neural systems, 13 (2002) 2, p. 195- 216 
DOI number (of the published article): 10.1088/0954-898X/13/2/303
Bibtex
Download full preprint: PDF (513 kB), PS ziped (477 kB)

Abstract:
The discrete-time dynamics of particular two and three neuron networks is studied with respect to non-trivial bifurcation scenarios. Interesting dynamical properties like periodic, quasi-periodic and chaotic attractors are observed, many of them coexisting for one and the same set of parameters. Symmetries in parameter space leading to equivalent dynamics are considered. Conditions on the connectivity of these small netwoks, involving excitation as well as inhibition, are derived which admit complex dynamics for these systems.