SPATIAL ORGANIZATION OF NEURAL NETWORKS: A PROBABILISTIC MODELING APPROACH.
Date
1987Publisher
IEEESource
1987 IEEE Conference on Neural Information Processing Systems - Natural and Synthetic.Google Scholar check
Keyword(s):
Metadata
Show full item recordAbstract
Summary form only given. A probabilistic model of neural networks was developed using the theory of general product-form queuing networks. A neural network is modeled as an open network of nodes, in which customers moving from node to node represent stimulation and connections between nodes are expressed in terms of suitably selected routing probabilities. A solution of the model is obtained under different disciplines affecting the time spent by a stimulation at each node visited. Results concerning the distribution of excitation as a function of network topology are compared with measures obtained by simulating the behavior of conventional neural networks.