Feature detection using spikes: the greedy approach
Abstract
A goal of lowlevel neural processes is to build an efficient code extracting the relevant information from the sensory input. It is believed that this is implemented in cortical areas by elementary inferential computations dynamically extracting the most likely parameters corresponding to the sensory signal. We explore here a neuromimetic feedforward model of the primary visual area (VI) solving this problem in the case where the signal may be described by a robust linear generative model. This model uses an overcomplete dictionary of primitives which provides a distributed probabilistic representation of input features. Relying on an efficiency criterion, we derive an algorithm as an approximate solution which uses incremental greedy inference processes. This algorithm is similar to 'Matching Pursuit' and mimics the parallel architecture of neural computations. We propose here a simple implementation using a network of spiking integrateandfire neurons which communicate using lateral interactions. Numerical simulations show that this Sparse Spike Coding strategy provides an efficient model for representing visual data from a set of natural images. Even though it is simplistic, this transformation of spatial data into a spatiotemporal pattern of binary events provides an accurate description of some complex neural patterns observed in the spiking activity of biological neural networks.
 Publication:

arXiv eprints
 Pub Date:
 November 2006
 arXiv:
 arXiv:qbio/0611003
 Bibcode:
 2006q.bio....11003P
 Keywords:

 Quantitative Biology  Neurons and Cognition
 EPrint:
 This work links Matching Pursuit with bayesian inference by providing the underlying hypotheses (linear model, uniform prior, gaussian noise model). A parallel with the parallel and eventbased nature of neural computations is explored and we show application to modelling Primary Visual Cortex / image processsing. http://incm.cnrsmrs.fr/perrinet/dynn/LaurentPerrinet/Publications/Perrinet04tauc