• DocumentCode
    2714825
  • Title

    Dimension reduction, generalized deformable models and the development of occularity and orientation

  • Author

    Yuille, A.L. ; Kolodny, J.A. ; Lee, C.W.

  • Author_Institution
    Div. of Appl. Sci., Harvard Univ., Cambridge, MA, USA
  • fYear
    1991
  • fDate
    8-14 Jul 1991
  • Firstpage
    597
  • Abstract
    It has been proposed that the spatial organization of cells in the visual cortex satisfies an abstract optimization criterion based on a smooth mapping criterion. The authors apply this criterion to the development of occularity and show that it can be expressed in terms of a generalized deformable model. From this model one can derive two alternative algorithms, one based on elastic nets and the other on Hebb rules. This shows a surprising mathematical relation between the elastic net model of G.J. Goodhill and D. Willshaw (1990) and the Hebbian models of K.D. Miller et al. (1989) and Tanaka (1990), even though their biophysical mechanisms are very different. Applying the same techniques to the development of orientation shows mathematically that a variant of the dimension reduction criterion of R. Durbin and G. Mitchison (1990) can be closely approximated by an elastic net. Computer simulations of the present algorithms give good agreement with experimental data
  • Keywords
    biocybernetics; brain models; eye; neural nets; neurophysiology; Hebb rules; abstract optimization criterion; biophysical mechanisms; computer simulations; dimension reduction; elastic nets; generalized deformable models; occularity; orientation; spatial organization of cells; Biological system modeling; Brain modeling; Computational biology; Computer simulation; Deformable models; Laboratories; Mathematical model; Neurons; Psychology; Wiring;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Neural Networks, 1991., IJCNN-91-Seattle International Joint Conference on
  • Conference_Location
    Seattle, WA
  • Print_ISBN
    0-7803-0164-1
  • Type

    conf

  • DOI
    10.1109/IJCNN.1991.155402
  • Filename
    155402