Canonical ensemble approach to graded-response perceptrons

dc.contributor.author	Bolle, Desire	pt_BR
dc.contributor.author	Erichsen Junior, Rubem	pt_BR
dc.date.accessioned	2014-08-19T02:10:10Z	pt_BR
dc.date.issued	1999	pt_BR
dc.identifier.issn	1063-651X	pt_BR
dc.identifier.uri	http://hdl.handle.net/10183/101308	pt_BR
dc.description.abstract	Perceptrons with graded input-output relations and a limited output precision are studied within the Gardner- Derrida canonical ensemble approach. Soft non-negative error measures are introduced allowing for extended retrieval properties. In particular, the performance of these systems for a linear (quadratic) error measure, corresponding to the perceptron (adaline) learning algorithm, is compared with the performance for a rigid error measure, simply counting the number of errors. Replica-symmetry-breaking effects are evaluated, and the analytic results are compared with numerical simulations. [S1063-651X(99)04503-1]	en
dc.format.mimetype	application/pdf
dc.language.iso	eng	pt_BR
dc.relation.ispartof	Physical Review. E, Statistical Physics, Plasmas, Fluids and Related Interdisciplinary Topics. New York. Vol. 59, no. 3, pt. B (Mar. 1999), p. 3386-3401	pt_BR
dc.rights	Open Access	en
dc.subject	Redes neurais	pt_BR
dc.subject	Quebra de simetria	pt_BR
dc.subject	Dinamica de rede	pt_BR
dc.subject	Simulação	pt_BR
dc.title	Canonical ensemble approach to graded-response perceptrons	pt_BR
dc.type	Artigo de periódico	pt_BR
dc.identifier.nrb	000055633	pt_BR
dc.type.origin	Estrangeiro	pt_BR

Este item está licenciado na Creative Commons License