ML Anthology

Neural Computation with Winner-Take-All as the Only Nonlinear Operation

Wolfgang Maass
NeurIPS 1999 pp. 293-299
/neurips/1999/maass1999neurips-neural/

Abstract

Everybody "knows" that neural networks need more than a single layer of nonlinear units to compute interesting functions. We show that this is false if one employs winner-take-all as nonlinear unit:

PDF NeurIPS Semantic Scholar

Cite

Text

Maass. "Neural Computation with Winner-Take-All as the Only Nonlinear Operation." Neural Information Processing Systems, 1999.

Markdown

[Maass. "Neural Computation with Winner-Take-All as the Only Nonlinear Operation." Neural Information Processing Systems, 1999.](https://mlanthology.org/neurips/1999/maass1999neurips-neural/)

BibTeX

@inproceedings{maass1999neurips-neural,
  title     = {{Neural Computation with Winner-Take-All as the Only Nonlinear Operation}},
  author    = {Maass, Wolfgang},
  booktitle = {Neural Information Processing Systems},
  year      = {1999},
  pages     = {293-299},
  url       = {https://mlanthology.org/neurips/1999/maass1999neurips-neural/}
}