Optimal signal representation in neural spiking codes: A model for the formation of simple cell receptive fields. - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2008

Optimal signal representation in neural spiking codes: A model for the formation of simple cell receptive fields.

Résumé

Taking advantage of the constraints of spiking representations, we derive an unsupervised learning algorithm solving efficiently the inverse problem of pattern matching and apply it to a model of the input to the primary visual cortex. In fact, spikes carry temporal event-based information in bundles of parallel fibers and may be considered as all-or-none binary events. This property may be used to formulate the efficiency of a representation problem as finding the L$_0$-norm sparsest representation, a ``hard" NP-complete problem. This framework improves previous results based on an Adaptive Matching Pursuit scheme by explicitly implementing an homeostatic constraint in the choice function by a spiking gain control mechanism in the neural population. For comparison purposes, we applied this scheme to the learning of small images taken from natural images as in SparseNet and compared the results and efficiency of this last algorithm with Matching Pursuit and the proposed algorithm. This study provides a generic algorithm for learning independent components in a set of inputs such as natural images suggesting that this Sparse Spike Coding strategy may provide a generic computational module that help us understanding the efficiency of the Primary Visual Cortex.
Fichier principal
Vignette du fichier
perrinet08shl.pdf (529.72 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-00156610 , version 1 (21-06-2007)
hal-00156610 , version 2 (05-09-2007)
hal-00156610 , version 3 (05-02-2008)
hal-00156610 , version 4 (18-03-2008)
hal-00156610 , version 5 (19-09-2008)
hal-00156610 , version 6 (25-06-2010)
hal-00156610 , version 7 (07-12-2016)

Identifiants

Citer

Laurent Perrinet. Optimal signal representation in neural spiking codes: A model for the formation of simple cell receptive fields.. 2008. ⟨hal-00156610v3⟩
299 Consultations
859 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More