NEURAL NETWORKS AND MATHEMATICAL DISTRIBUTIONS THEORY - PERIODICAL AND ALMOST PERIODICAL DIRAC FIRINGS

In this paper, we study a neural network. In this set, each neuron is characterized by its neural potential and its sequence of firings (Dir ac sequence distribution). All the neurons are connected by continuous links and also Dirac excitation (impulses = firings). With the use of a small parameter lambda, the neural network can be represented by a system of nonlinear differential equations. The existence of almost ge neralized periodical oscillation is established. The set has a memory, according to the theory of J.J. Hopfield [1].