Advertisement

Biological Cybernetics

, Volume 70, Issue 1, pp 65–73 | Cite as

Backpropagation learns Marr's operator

  • Anupam Joshi
  • Chia-Hoang Lee
Article

Abstract

This paper describes a neural network model of the retinal responses to stimuli whose architecture is inspired by neurophysiological data. Suitable assumptions are identified which enable the development of a simple model for an individual X-type ganglion cell using backpropagation. This is then used to make a model of retinal processing. We present here our model of the individual ganglion cells and the underlying assumptions. We show that backpropagation leads to a model which is similar to the mathematical descriptions of retinal processing advanced by Marr. We present the results obtained when our model is used to simulate the effect of retinal processing on images. Empirical results about the speedups obtained when this model is implemented on parallel architectures are also reported.

Keywords

Neural Network Simple Model Network Model Empirical Result Ganglion Cell 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Barrow HG (1987) Learning receptive fields. In: Proc IEEE 1st Int Conf on Neural networks (vol 4). pp 115–121Google Scholar
  2. Baylor DA, Fuortes MGF, O'Bryan PM (1971) Receptive fields of single cones in the retina of turtle. J Physiol (Lond) 214:256–294Google Scholar
  3. Braham R, Hamblen JO (1990) The design of a neural network with biologically motivated architecture. IEEE Trans Neural Networks 1:251–262Google Scholar
  4. Buser P, Imbert M (1992) Kay RH, (trans). Vision. MIT Press, Cambridge, MassGoogle Scholar
  5. Caldwell JH, Daw NW (1978) Effects of picrotoxin and strychnine on rabbit retinal ganglion cells: changes in centre surround receptive fields. J Physiol (Lond) 276:299–310Google Scholar
  6. Daugman J (1983) Six formal properties of two dimensional anisotropic visual filters. IEEE Trans Syst Man Cybern 13:882–887Google Scholar
  7. Daugman J (1988) Pattern and motion vision without Laplacian zero crossings. J Opt Soc Am [A] 5:1142–1148Google Scholar
  8. Dowling JE (1987) Retina, the approachable part of the brain. Harvard University Press, Cambridge, MassGoogle Scholar
  9. Enroth-Cugell C, Robson J (1966) The contrast sensitivity of retinal ganglion cells of the cat. J Physiol (Lond) 187:517–522Google Scholar
  10. Grossberg S, Wyse L (1991) Neural network architecture for figure-ground separation of connected scenic figures. Neural Networks 4:723–742Google Scholar
  11. Hubel DO (1988) Eye, brain and vision. Scientific American, New YorkGoogle Scholar
  12. Joshi A, Lee CH (1991) On modelling the retina using neural networks. In: Proc Int Joint Conf on Neural Networks'91, Singapore, pp 2343–2348Google Scholar
  13. Kleinschmidt J, Dowling JE (1975) Intracellular recordings from gecko photoreceptors during light and dark adaptation. J Gen Physiol 66:617–648aGoogle Scholar
  14. Kolb H (1979) The inner plexiform layer in the retina of the cat: electron microscopic observations. J Neurocytol 8:295–329Google Scholar
  15. Lehky SR, Puget A, Sejnowski TJ (1990) Neural models of binocular depth perception. Cold Spring Harbor Symp Quant Biol 55:765–777Google Scholar
  16. Linsker R (1988) Self organisation in a perceptual network. Computer 21:105–117Google Scholar
  17. Mach E (1906) Über den Einfluss räumlich und zeitlich variieender Lichtreize auf die Gesichtswahrnehmung. Sitzungsber Math Naturwiss Kl Kaiser Akad Wiss 115:633–648Google Scholar
  18. Marr D (1982) Vision. Freeman, San FranciscoGoogle Scholar
  19. Marr D, Hildreth E (1980) The theory of edge detection. Proc R Soc London Ser B 207:187–217Google Scholar
  20. Oğuztöreli M (1988) Underlying neural computations for some visual phenomenon. Biol Cybern 60:89–106Google Scholar
  21. Oğuztöreli M (1989) Neural computations in some visual processes. In: Proc 1989 IEEE Int Conf on Systems, Man and Cybernetics, vol 2, pp 664–670Google Scholar
  22. Richter J, Ullman S (1986) Non linearities in cortical simple cells and the possible detection of zero crossing. Biol Cybern 53:195–202Google Scholar
  23. Thibos LN, Werblin FS (1978) The response properties of the steady antagonistic surround in the mudpuppy retina. J Physiol (Lond) 278:79–99Google Scholar
  24. Wallach H (1948) Brightness constancy and the nature of achromatic colours. J Exp Psychol 38:310–324Google Scholar
  25. Watt R (1988) Visual processing: computational psychophysics and cognitive research. Erlbaum, HillsdaleGoogle Scholar
  26. West RW (1976) Light and electron microscopy of the ground squirrel retina: functional considerations. J Comp Neurol 168:355–377Google Scholar
  27. West RW, Dowling JE (1972) Synapses onto different morphological types of retinal ganglion cells. Science 178:510–512Google Scholar

Copyright information

© Springer-Verlag 1993

Authors and Affiliations

  • Anupam Joshi
    • 1
  • Chia-Hoang Lee
    • 2
  1. 1.Department of Computer SciencePurdue UniversityWest LafayetteUSA
  2. 2.Department of Computer and Information ScienceNational Chiao-Tung UniversityHsinchu, TaiwanRepublic of China

Personalised recommendations