Decreasing the size of the Restricted Boltzmann machine
Machine Learning
2018-12-13 v2 Machine Learning
Abstract
We propose a method to decrease the number of hidden units of the restricted Boltzmann machine while avoiding decrease of the performance measured by the Kullback-Leibler divergence. Then, we demonstrate our algorithm by using numerical simulations.
Cite
@article{arxiv.1807.02999,
title = {Decreasing the size of the Restricted Boltzmann machine},
author = {Yohei Saito and Takuya Kato},
journal= {arXiv preprint arXiv:1807.02999},
year = {2018}
}