Revista: | Brazilian journal of physics |
Base de datos: | PERIÓDICA |
Número de sistema: | 000146995 |
ISSN: | 0103-9733 |
Autores: | Diambra, L1 |
Instituciones: | 1Universidade de Sao Paulo, Instituto de Física, Sao Paulo. Brasil |
Año: | 1998 |
Periodo: | Dic |
Volumen: | 28 |
Número: | 4 |
Paginación: | 453-461 |
País: | Brasil |
Idioma: | Inglés |
Tipo de documento: | Artículo |
Enfoque: | Analítico |
Resumen en inglés | We investigate the learning of a rule from examples of the case of boolean perceptron. Previous studies of this problem have been made using the full quenched theory. We consider here two alternative approaches that can be applied easily. the two-replicas interactions approach considerably improves upon the well-known first-order approach. The mean field approach proved some results that have been obtained previously using the complex full quenched theory. Both approximations have been applied to both continuous weights and discrete weights perceptron |
Disciplinas: | Física y astronomía |
Palabras clave: | Física, Termodinámica y física estadística, Sistemas complejos, Redes neuronales, Aprendizaje, Perceptrón booleano |
Keyword: | Physics and astronomy, Physics, Thermodynamics and statistical physics, Complex systems, Neural networks, Learning, Boolean perceptron |
Texto completo: | Texto completo (Ver HTML) |