The Ni1000: High Speed Parallel VLSI for Implementing Multilayer Perceptrons

Abstract

In this paper we present a new version of the standard multilayer perceptron (MLP) algorithm for the state-of-the-art in neural net(cid:173) work VLSI implementations: the Intel Ni1000. This new version of the MLP uses a fundamental property of high dimensional spaces which allows the 12-norm to be accurately approximated by the It -norm. This approach enables the standard MLP to utilize the parallel architecture of the Ni1000 to achieve on the order of 40000, 256-dimensional classifications per second.

Cite

Text

Perrone and Cooper. "The Ni1000: High Speed Parallel VLSI for Implementing Multilayer Perceptrons." Neural Information Processing Systems, 1994.

Markdown

[Perrone and Cooper. "The Ni1000: High Speed Parallel VLSI for Implementing Multilayer Perceptrons." Neural Information Processing Systems, 1994.](https://mlanthology.org/neurips/1994/perrone1994neurips-ni1000/)

BibTeX

@inproceedings{perrone1994neurips-ni1000,
  title     = {{The Ni1000: High Speed Parallel VLSI for Implementing Multilayer Perceptrons}},
  author    = {Perrone, Michael P. and Cooper, Leon N.},
  booktitle = {Neural Information Processing Systems},
  year      = {1994},
  pages     = {747-754},
  url       = {https://mlanthology.org/neurips/1994/perrone1994neurips-ni1000/}
}