Emergence of filters from natural scenes in a sparse spike coding scheme
First detailed report on learning Receptive Fields using spiking neurons which ressembles RFs of simple cells in Primary Visual Cortex. This followed a first report in Publications/Perrinet02sparse. It also extended the learning to a topological organization of cells, creating a model for pin-wheel formation (see Fig. 1) This work was continued in
a comparison with SparseNet (see Publications/Perrinet06cns)
a complete code architecture (see SparseHebbianLearning)
|
Figure 1 : Introducing lateral connections between filters in a model of learning (see Publications/Perrinet02sparse) in favor of neighboring activity, we observe the emergence of topological relations between neighboring filters (cyclically numbered from 1 to 16 from left to right and then from bottom to down) similar to the neurophysiologically observed pinwheels. |
reference
- Laurent Perrinet, Manuel Samuelides, Simon Thorpe. Emergence of filters from natural scenes in a sparse spike coding scheme., URL . Neurocomputing, 58--60(C):821--6, 2003 abstract.
All material (c) L. Perrinet. Please check the copyright notice.