A novel chaotic neural network architecture.
Nigel T. Crook, Tjeerd V. olde Scheper · 2001
The basic premise of this research is that deterministic chaos is a powerful mechanismfor the storage and retrieval of information in the dynamics of artificial neuralnetworks. Substantial evidence has been found in biological studies for the presenceof chaos in the dynamics of natural neuronal systems [1-3]. Many have suggestedthat this chaos plays a central role in memory storage and retrieval [1,4-6]. Indeed,chaos offers many advantages over alternative memory storage mechanisms used inartificial neural networks. One is that chaotic dynamics are significantly easier tocontrol than other linear or non-linear systems, requiring only small appropriatelytimed perturbations to constrain them within specific Unstable Periodic Orbits(UPOs). Another is that chaotic attractors contain an infinite number of these UPOs.If individual UPOs can be made to represent specific internal memory states of asystem, then in theory a chaotic attractor can provide an infinite memorystore for thesystem. In this paper we investigate the possibility that a network can self-selectUPOs in response to specific dynamic input signals. These UPOs correspond tonetwork recognition states for these input signals.