Microelectronic Implementations of Connectionist Neural Networks
Abstract
In this paper we discuss why special purpose chips are needed for useful implementations of connectionist neural networks in such applications as pattern recognition and classification. Three chip designs are described: a hybrid digital/analog programmable connection matrix, an analog connection matrix with adjustable connection strengths, and a digital pipe lined best-match chip. The common feature of the designs is the distribution of arithmetic processing power amongst the data storage to minimize data movement.
Cite
Text
Mackie et al. "Microelectronic Implementations of Connectionist Neural Networks." Neural Information Processing Systems, 1987.Markdown
[Mackie et al. "Microelectronic Implementations of Connectionist Neural Networks." Neural Information Processing Systems, 1987.](https://mlanthology.org/neurips/1987/mackie1987neurips-microelectronic/)BibTeX
@inproceedings{mackie1987neurips-microelectronic,
title = {{Microelectronic Implementations of Connectionist Neural Networks}},
author = {Mackie, Stuart and Graf, Hans Peter and Schwartz, Daniel B. and Denker, John S.},
booktitle = {Neural Information Processing Systems},
year = {1987},
pages = {515-523},
url = {https://mlanthology.org/neurips/1987/mackie1987neurips-microelectronic/}
}