Turing universality of neural nets (revisited)

Título:	Turing universality of neural nets (revisited)
Autores/as:	Neto, Pedro J. Siegelmann, Hava T. Costa, Félix J. Suárez Araujo, C. P.
Clasificación UNESCO:	120304 Inteligencia artificial
Palabras clave:	Modularity Neural computation Recursive function theory
Fecha de publicación:	1997
Publicación seriada:	Lecture Notes in Computer Science
Conferencia:	6th International Workshop on Computer Aided Systems Theory, EUROCAST 1997
Resumen:	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.
URI:	http://hdl.handle.net/10553/73316
ISBN:	3-540-63811-3
ISSN:	0302-9743
DOI:	10.1007/BFb0025058
Fuente:	Pichler F., Moreno-Díaz R. (eds) Computer Aided Systems Theory — EUROCAST 1997. Lecture Notes in Computer Science, v. 1333, p. 361-366. Springer, Berlin, Heidelberg [ISSN 0302-9743], (Enero 1997)
Colección:	Actas de congresos

Vista completa