Non-Linear Dimensionality Reduction

Abstract

A method for creating a non-linear encoder-decoder for multidimensional data with compact representations is presented. The commonly used technique of autoassociation is extended to allow non-linear representations, and an objec(cid:173) tive function which penalizes activations of individual hidden units is shown to result in minimum dimensional encodings with respect to allowable error in reconstruction.

Cite

Text

DeMers and Cottrell. "Non-Linear Dimensionality Reduction." Neural Information Processing Systems, 1992.

Markdown

[DeMers and Cottrell. "Non-Linear Dimensionality Reduction." Neural Information Processing Systems, 1992.](https://mlanthology.org/neurips/1992/demers1992neurips-nonlinear/)

BibTeX

@inproceedings{demers1992neurips-nonlinear,
  title     = {{Non-Linear Dimensionality Reduction}},
  author    = {DeMers, David and Cottrell, Garrison W.},
  booktitle = {Neural Information Processing Systems},
  year      = {1992},
  pages     = {580-587},
  url       = {https://mlanthology.org/neurips/1992/demers1992neurips-nonlinear/}
}