A Method for the Efficient Design of Boltzmann Machines for Classiffication Problems
Abstract
We introduce a method for the efficient design of a Boltzmann machine (or a Hopfield net) that computes an arbitrary given Boolean function f . This method is based on an efficient simulation of acyclic circuits with threshold gates by Boltzmann machines. As a consequence we can show that various concrete Boolean functions f that are relevant for classification problems can be computed by scalable Boltzmann machines that are guaranteed to converge to their global maximum configuration with high probability after constantly many steps.
Cite
Text
Gupta and Maass. "A Method for the Efficient Design of Boltzmann Machines for Classiffication Problems." Neural Information Processing Systems, 1990.Markdown
[Gupta and Maass. "A Method for the Efficient Design of Boltzmann Machines for Classiffication Problems." Neural Information Processing Systems, 1990.](https://mlanthology.org/neurips/1990/gupta1990neurips-method/)BibTeX
@inproceedings{gupta1990neurips-method,
title = {{A Method for the Efficient Design of Boltzmann Machines for Classiffication Problems}},
author = {Gupta, Ajay and Maass, Wolfgang},
booktitle = {Neural Information Processing Systems},
year = {1990},
pages = {825-831},
url = {https://mlanthology.org/neurips/1990/gupta1990neurips-method/}
}