Simulations with a large number of neurons
T. J. P. Penna, P.M.C. de Oliveira · Journal of Physics A Mathematical and General · 1989
The authors propose a new algorithm for evolution of neural networks, by division of the network into subnets, in order to reach retrieval of stored correlated patterns. They present a fast code for simulations with a large number of neurons ( approximately 10000). The simulations were carried out on a microcomputer.