Optimally adapted multistate neural networks trained with noise

The principle of adaptation in a noisy retrieval environment is extended here to a diluted attractor neural network of Q-state neurons trained with noisy data. The network is adapted to an appropriate noisy training overlap and training activity, which are determined self-consistently by the optimiz...

Descripción completa

Detalles Bibliográficos
Autores: Erichsen Junior, Rubem, Theumann, Walter Karl
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:1999
País:Brasil
Institución:Universidade Federal do Rio Grande do Sul (UFRGS)
Repositorio:Repositório Institucional da UFRGS
Idioma:inglés
OAI Identifier:oai:www.lume.ufrgs.br:10183/103647
Acceso en línea:http://hdl.handle.net/10183/103647
Access Level:acceso abierto
Palabra clave:Física estatística
Redes neurais
Biofísica
Transformacoes de ordem-desordem
Descripción
Sumario:The principle of adaptation in a noisy retrieval environment is extended here to a diluted attractor neural network of Q-state neurons trained with noisy data. The network is adapted to an appropriate noisy training overlap and training activity, which are determined self-consistently by the optimized retrieval attractor overlap and activity. The optimized storage capacity and the corresponding retriever overlap are considerably enhanced by an adequate threshold in the states. Explicit results for improved optimal performance and new retriever phase diagrams are obtained for Q=3 and Q=4, with coexisting phases over a wide range of thresholds. Most of the interesting results are stable to replica-symmetry-breaking fluctuations.