A network that uses few active neurones to code visual input predicts the diverse shapes of cortical receptive fields

@article{Rehn2006ANT,
  title={A network that uses few active neurones to code visual input predicts the diverse shapes of cortical receptive fields},
  author={Martin Rehn and Friedrich T. Sommer},
  journal={Journal of Computational Neuroscience},
  year={2006},
  volume={22},
  pages={135-146}
}
  • Martin Rehn, F. Sommer
  • Published 15 February 2007
  • Biology, Computer Science
  • Journal of Computational Neuroscience
Computational models of primary visual cortex have demonstrated that principles of efficient coding and neuronal sparseness can explain the emergence of neurones with localised oriented receptive fields. Yet, existing models have failed to predict the diverse shapes of receptive fields that occur in nature. The existing models used a particular “soft” form of sparseness that limits average neuronal activity. Here we study models of efficient coding in a broader context by comparing soft and… 

Figures from this paper

A Sparse Coding Model with Synaptically Local Plasticity and Spiking Neurons Can Account for the Diverse Shapes of V1 Simple Cell Receptive Fields

TLDR
It is proved, mathematically, that sparseness and decorrelation are the key ingredients that allow for synaptically local plasticity rules to optimize a cooperative, linear generative image model formed by the neural representation.

Sparse coding models predict a spectral bias in the development of primary visual cortex (V1) receptive fields

TLDR
This work trains an overcomplete sparse coding model (Sparsenet) on natural images and finds that there is indeed order in the development of its basis functions, with basis functions tuned to lower spatial frequencies emerging earlier and higher spatial frequency basis functions emerging later.

Complex Sparse Code Priors Improve the Statistical Models of Neurons in Primate Primary Visual Cortex

TLDR
It is found that models with complex sparse code prior are significantly better than models with classical Gabor and/or Laplacian of Gaussian filter priors for modeling and predicting the activities of neurons that have been earlier identified to exhibit higher-order (HO) complex pattern selectivity.

Inhibitory Interneurons Decorrelate Excitatory Cells to Drive Sparse Code Formation in a Spiking Model of V1

TLDR
It is shown that adding a separate population of inhibitory neurons to a spiking model of V1 provides conformance to Dale's Law, proposes a computational role for at least one class of interneurons, and accounts for certain observed physiological properties in V1.

Role of Homeostasis in Learning Sparse Representations

TLDR
By quantitatively assessing the efficiency of the neural representation during learning, a cooperative homeostasis mechanism is derived that optimally tunes the competition between neurons within the sparse coding algorithm.

Sparse Coding Can Predict Primary Visual Cortex Receptive Field Changes Induced by Abnormal Visual Input

TLDR
In every condition, the changes in receptive field properties previously observed experimentally were matched to a similar and highly faithful degree by all the models, suggesting that early sensory development can indeed be understood in terms of an impetus towards sparsity.

Constrained inference in sparse coding reproduces contextual effects and predicts laminar neural dynamics

TLDR
An extended generative model for visual scenes that includes spatial dependencies among different features and provides a novel framework for contextual processing in the visual system proposing a well-defined functional role for horizontal axons and feedback projections is proposed.

Perceptual Decision Making “Through the Eyes” of a Large-Scale Neural Model of V1

TLDR
It is shown that while both sparse and non-sparse linear decoding yield discrimination results quantitatively consistent with human psychophysics, sparse linear decoding is more efficient in terms of the number of selected informative dimension.

Learning modular representations from global sparse coding networks

TLDR
It is suggested that sparse coding must be refined to incorporate stimulus-dependent plasticity, and as connections are strengthened amongst all the cells that become active in response to a given stimuli, the networks that emerge exhibit modularity and small-world topologies, both of which have been observed in networks in the cortex.

Sparse Coding Models Can Exhibit Decreasing Sparseness while Learning Sparse Codes for Natural Images

TLDR
It is demonstrated that some sparse coding models, such as those employing homeostatic mechanisms on neural firing rates, can exhibit decreasing sparseness during learning, while still achieving good agreement with mature V1 receptive field shapes and a reasonably sparse mature network state.
...

References

SHOWING 1-10 OF 63 REFERENCES

Emergence of simple-cell receptive field properties by learning a sparse code for natural images

TLDR
It is shown that a learning algorithm that attempts to find sparse linear codes for natural scenes will develop a complete family of localized, oriented, bandpass receptive fields, similar to those found in the primary visual cortex.

Spatial structure and symmetry of simple-cell receptive fields in macaque primary visual cortex.

TLDR
Measurements of the spatial structure of simple-cell receptive fields in macaque primary visual cortex show a tendency for neurons that are well tuned in orientation and spatial frequency to have odd-symmetric receptive fields, and a population analysis reveals that the distribution of spatial profiles inPrimary visual cortex lies approximately on a one-parameter family of filter shapes.

Sparse coding in the primate cortex

TLDR
There are now strong theoretical reasons and experimental evidence suggesting that the brain adopts a compromise between these extremes which is often referred to as sparse coding.

Simple-Cell-Like Receptive Fields Maximize Temporal Coherence in Natural Video

TLDR
It is shown that simple-cell-type receptive fields emerge when temporal response strength correlation is maximized for natural image sequences, and suggests an interpretation of simple cells in terms of invariant coding principles, which have previously been used to explain complex-cell receptive fields.

The “independent components” of natural scenes are edge filters

Sparse coding with an overcomplete basis set: A strategy employed by V1?

The two-dimensional spatial structure of simple receptive fields in cat striate cortex.

TLDR
A reverse correlation method is developed that allows quantitative determination of visual receptive-field structure in two spatial dimensions and it is demonstrated that thereverse correlation method yields results with several desirable properties, including convergence and reproducibility independent of modest changes in stimulus parameters.

Toward a Theory of the Striate Cortex

We explore the hypothesis that linear cortical neurons are concerned with building a particular type of representation of the visual worldone that not only preserves the information and the

Towards a theory of striate cortex

We explore the hypothesis that linear cortical neurons are concerned with building a particular type of representation of the visual world --- one which not only preserves the information and the
...