Dimensionality-Reduction Using Connectionist Networks

  title={Dimensionality-Reduction Using Connectionist Networks},
  author={Eric Saund},
  journal={IEEE Trans. Pattern Anal. Mach. Intell.},
A method is presented for using connectionist networks of simple computing elements to discover a particular type of constraint in multidimensional data. Suppose that some data source provides samples consisting of n-dimensional feature-vectors, but that this data all happens to lie on an rn-dimensional surface embedded in the k-dimensional feature space. Then occurrences of data can be more concisely described by specifying an rn-dimensional location on the embedded surface than by reciting… CONTINUE READING


Publications referenced by this paper.
Showing 1-10 of 10 references

Application of the Karhunen - Lokve expansion to feature selection and ordering , ” IEEE Trans . Comput . , S . Hanson and D . Burr , “ Knowledge representation in connectionist networks

  • T. Sejnowski G. Hinton, D. Ackley
  • 1987

McClelland, and the PDP Research Group, Parallel Distributed Processing: Explorations in the Structure of Cognition

  • J. D. Rumelhart
  • 1986


  • Biol. Cybern..
  • Kirkpatrick, S. Gelatt, and M. Vecchi…
  • 1985

Heydorn, “Some optimal approaches to feature extraction,’

  • R. I . Tou
  • Computer and Information Sciences 11,
  • 1967

Learning internal representations by error propagation , ” ICs Rep . 8506 , Instit . Cognitive Sci . , UCSD , 1985 . - , “ Learning internal representations by back - propagating errors , ”

  • R. Williams
  • Nature
  • 1962

Similar Papers

Loading similar papers…