Differences between revisions 29 and 30
Revision 29 as of 2012-01-08 16:30:05
Size: 15
Comment:
Revision 30 as of 2013-09-12 07:50:22
Size: 1624
Comment:
Deletions are marked like this. Additions are marked like this.
Line 2: Line 2:

  = Sparse spike coding in an asynchronous feed-forward multi-layer neural network using Matching Pursuit =

This paper introduced for the first time the use of Matching Pursuit in a neuromimetic architecture of low-level visual areas. It aims at understanding the efficiency and generality of correlation-based inhibition propagation, by showing:

 * the link between sparse coding, spike coding and matching pursuit,
 * the use of a quantization in Matching Pursuit (see [[Figures/Perrinet03ieee/FigureLut| Fig. 1]]),
 * the efficiency of this algorithm (e.g. for compression as compared with JPEG see [[Figures/Perrinet02sparse/FigureDeux| Fig. 2]]) -- this was developped in [[Publications/Perrinet03ieee| (Perrinet, 03, IEEE)]] and [[Publications/Fischer07| (Fischer, 07)]]),
 * that a [[SparseHebbianLearning| Sparse Hebbian Learning]] scheme could be applied, leading to the learning of edge-like filter (this is the first report of using an adaptive matching pursuit scheme to my knowledge) -- this was developped in [[Publications/Perrinet03ieee| (Perrinet, 03, IEEE)]] and [[Publications/Fischer07| (Fischer, 07)]]), 
 * the possible use as a model of low-level saliency.
See also 
 * [[Publications/Perrinet03ieee]]
 * [[Publications/Perrinet02esann]]
## TODO fix: || <<Include(Figures/Perrinet02sparse/FigureUn)>> || <<Include(Figures/Perrinet02sparse/FigureDeux)>> ||
##<<Include(Figures/Perrinet02sparse/FigureDeux)>>
 * Get a [[attachment:perrinet02sparse.pdf|preprint]]

## * see a [[attachment:../Perrinet02esann/perrinet02sparse_presentation.pdf|presentation]]

Sparse spike coding in an asynchronous feed-forward multi-layer neural network using Matching Pursuit

This paper introduced for the first time the use of Matching Pursuit in a neuromimetic architecture of low-level visual areas. It aims at understanding the efficiency and generality of correlation-based inhibition propagation, by showing:

  • the link between sparse coding, spike coding and matching pursuit,
  • the use of a quantization in Matching Pursuit (see Fig. 1),

  • the efficiency of this algorithm (e.g. for compression as compared with JPEG see Fig. 2) -- this was developped in (Perrinet, 03, IEEE) and (Fischer, 07)),

  • that a Sparse Hebbian Learning scheme could be applied, leading to the learning of edge-like filter (this is the first report of using an adaptive matching pursuit scheme to my knowledge) -- this was developped in (Perrinet, 03, IEEE) and (Fischer, 07)), 

  • the possible use as a model of low-level saliency.

See also 

welcome: please sign in