Skip to Main content Skip to Navigation

Modèles Impulsionnels de Réseaux de Neurones Biologiques

Abstract : The classical point of view in computational neuroscience is that neurons process and code information with firing rates. Recently, experimental evidence of neural synchronization at a fine time scale has motivated a renewed interest in the time coding theory, according to which neurons process information with precisely timed spike trains. Since spiking models are not as well-understood as rate models, we start by establishing a body of general mathematical results, which allow us to show that these models respond reliably to aperiodic stimuli. Thanks to this property, we design a spiking model of orientation selectivity in the primary visual cortex, the spiking perceptron. Because the model detects a geometrical property of the image instead of matching the stimulus with a template, it responds naturally in a contrast-invariant way with a feedforward architecture.
Document type :
Complete list of metadatas

Cited literature [201 references]  Display  Hide  Download
Contributor : Romain Brette <>
Submitted on : Saturday, March 13, 2004 - 3:13:38 PM
Last modification on : Wednesday, December 9, 2020 - 3:09:17 PM
Long-term archiving on: : Friday, April 2, 2010 - 7:52:51 PM


  • HAL Id : tel-00005340, version 1


Romain Brette. Modèles Impulsionnels de Réseaux de Neurones Biologiques. Neurosciences [q-bio.NC]. Université Pierre et Marie Curie - Paris VI, 2003. Français. ⟨tel-00005340⟩



Record views


Files downloads