740 SPATIAL ORGANIZATION OF NEURAL NEn

The aim of this paper is to explore the spatial organization of neural networks under Markovian assumptions, in what concerns the behaviour of individual cells and the interconnection mechanism. Spaceorganizational properties of neural nets are very relevant in image modeling and pattern analysis, where spatial computations on stochastic two-dimensional image fields are involved. As a first approach we develop a random neural network model, based upon simple probabilistic assumptions, whose organization is studied by means of discrete-event simulation. We then investigate the possibility of approXimating the random network's behaviour by using an analytical approach originating from the theory of general product-form queueing networks. The neural network is described by an open network of nodes, in which customers moving from node to node represent stimulations and connections between nodes are expressed in terms of suitably selected routing probabilities. We obtain the solution of the model under different disciplines affecting the time spent by a stimulation at each node visited. Results concerning the distribution of excitation in the network as a function of network topology and external stimulation arrival pattern are compared with measures obtained from the simulation and validate the approach followed.