Title :
Chaos in neural networks with bistable dendrite membrane current-voltage relation
Author :
Gartliauskas, A.
Author_Institution :
Lab. of Neuroinf., Inst. of Math. & Inf., Vilnius
Abstract :
The study of a chaos theory, in general, and a neural network chaos paradigm with a concrete interpretation by a simple neural network, in particular, allowed us to set up a new aspect in an artificial neuronal methodology: an analogy between natural chaos experimentally observed in neural systems of the brain and artificial neural network chaos phenomena is considered. The significance of asymmetry and nonlinearity, which were increased on introducing a restricted N-shaped synapsed relation in a dynamic model, is emphasized, There are illustrated the computational examples of the neural network properties which are expressed by the equilibrium point, stable cycle or chaotic behaviour in neural systems with N-shaped synapsed relation
Keywords :
bioelectric phenomena; biomembranes; brain models; chaos; neural nets; neurophysiology; asymmetry; bistable dendrite membrane current-voltage relation; brain; dynamic model; equilibrium point; neural network chaos paradigm; nonlinearity; restricted N-shaped synapsed relation; stable cycle; Artificial neural networks; Biological neural networks; Biomembranes; Chaos; Computer networks; Humans; Intelligent networks; Neural networks; Orbits; Pattern recognition;
Conference_Titel :
Neural Networks, 1999. IJCNN '99. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-5529-6
DOI :
10.1109/IJCNN.1999.831593