We show how to use recursive function theory to prove Turing universality of finite analog recurrent neural nets, with a piecewise linear sigmoid function as activation function. We emphasize the modular construction of nets within nets, a relevant issue from the software engineering point of view.
Keywords Neural computation - recursive function theory - modularity
This work was supported by JNICT PBIC/TIT/2527/95 and a fellowship from the Gobierno Autonomo de Canarias.