BibTeX
@MISC{Stollenwerk_switchingin,
author = {Nico Stollenwerk and Frank Pasemann},
title = {Switching in Self-Controlled Chaotic},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
The activity dynamics of recurrent neural networks can exhibit deter-ministic chaos due to the nonlinear transfer functions. Chaotic attractors are winded around infinitely many unstable periodic orbits. Each can be stabilized by a feedback control. In the present article we explore the flex-ibility of the chaotic dynamics of a recurrent neuromodule and construct a neural controller which is able to switch between several periodic pat-terns in two different ways: either deterministically by external inputs or spontaneously by dynamic noise. The chaotic attractor acts as an interme-diate state between successively stabilized dynamic patterns. It has all the possible motions present, like an attentive state between different stimuli.
Keyphrases
self-controlled chaotic deter-ministic chaos external input dynamic pattern present article chaotic attractor recurrent neural network different stimulus recurrent neuromodule different way dynamic noise neural controller attentive state chaotic dynamic several periodic pat-terns interme-diate state feedback control activity dynamic nonlinear transfer function many unstable periodic orbit chaotic attractor act possible motion
