| Autor: |
Duch, Włodzisław, Kacprzyk, Janusz, Oja, Erkki, Zadrożny, Sławomir, Araujo, C. P. Suárez, López, P. Fernández, Báez, P. García, García, J. Regidor |
| Zdroj: |
Artificial Neural Networks: Biological Inspirations - ICANN 2005; 2005, p247-253, 7p |
| Abstrakt: |
The Computational Neuroscience has as main goal the understanding of the computational style of the brain and developing artificial systems with brain capabilities. Our paper belongs to this field. We will use an Hebbian neural ensemble which follow a non-linear differential equation system namely Hebbian System (HS), which represent the neurodynamics and the adaptation in accordance with the Hebb's postulate, to study the influence of the NO diffusion in the Hebbian learning. Considering that the postsynaptic neurons provide retrograde signals to the presynaptic neurons [1] we suggest the NO as a probable biological support to the Hebb's law propounding a new mathematical formulation of that learning law, the diffusive Hebb's law. We will present a study of the behavior of the diffusive Hebb's law using a Diffusive Hebbian System (DHS). [ABSTRACT FROM AUTHOR] |
| Databáze: |
Supplemental Index |
| Externí odkaz: |
|
