Show Reference: "Natural Image Statistics and Neural Representation"

See the CiteULike entry for more info, PDF links, BibTex etc.

The visual cortex is hierarchically organized.

The idea that neural activity does not primarily represent the world but 'action pointers', as put by Engel et al., speaks to the deep SC which is both 'multi-modal' and 'motor'.

The `efficient coding principle' states that a neural ensemble should encode as much information as possible in its response.

To estimate optimally, it is necessary to take into account the rate of each stimulus value. This is neglected by the efficient coding approach, which is recognized by the opponents.

The pure efficient coding hypothesis does not take into account noise which may corrupt signals.

The efficient coding hypothesis does not take into account any task neural processing is supposed to accomplish. Some redundancy may make a code more suitable for a particular task. This is true especially when the values being represented are not equally distributed, when there is noise, and when responding correctly to some values yields higher utility than for others.

"Natural images are statistically redundant."

In an efficient population code, neural responses are statistically independent.

It seems as though the primates' trichromatic visual system is well-suited to capture the distribution of colors in natural systems.

By optimizing sparseness (or coding efficiency) of functions for representing natural images, one can arrive at tuning functions similar to those found in in simple cells. They are

  • spatially localized
  • oriented
  • band-pass filters with different spatial frequencies.

LGN cells respond whitened---ie. efficiently---to natural images, but they respond non-white to white noise, eg. They are thus well-adapted to natural images from the efficient coding point of view.

One hypothesis about early visual processing is that it tries to preserve (and enhance) as much information about the visual stimuli (with as little effort) as possible. Findings about efficiency in visual processing seem to validate this hypothesis.

A complete theory of early visual processing would need to address more aspects than coding efficiency, optimal representation and cleanup. Tasks and implementation would have to be taken into account.