Random Recurrent Neural Networks for Autonomous System Design
Emmanuel Daucé, Mathias Quoy, Avenue Edouard Belin · 2000
In this article, we stress the need for using dynamical systems properties in autonomous architecture design. We first study the dynamics of random recurrent neural networks (RRNN). Such systems are known to spontaneously exhibits various dynamical regimes, as they always tries to remain on an attractor, thus achieving stable dynamical behaviors. Second, we try to characterize the adaptive properties of such a system in an open environment, i.e. in a system which always interacts with external signals.Under these conditions, a change in the behavior corresponds to the switch from one attractor to another one. Such bifurcation occur for very little changes in the environment signal; our system is thus unstable on its inputs. We propose a local Hebbian learning rule which tends to stabilize the response of the system for given inputs. After training, the system is able to perform recognition, i.e to produce a specific regular cyclic attractor while the learned input ...