Título: Formulation of Reference Field in Neural Networks with External Input: Mechanism for Sensory Feature Integration
Autores: Mogi, Kenichiro
Resumo: I aplly the graphic transformation method (K. Mogi, Phys. Rev. E 49, 4616 (1994)) to a neural network which receive input from, external neurons. I show that the state distribution of the neural network receiving input from the external neurons can be expressed as a product of the “reference field”, which is formed by the interaction between the external and “internal” neurons. The visual system can be described as a direct product of the reference field and the feature space.
Código DOI: 10.21528/CBRN1994-036
Artigo em PDF: CBRN1994-paper36.pdf
Arquivo BibTex: CBRN1994-paper36.bib