Sampling the RiemannTheta Boltzmann Machine
Abstract
We show that the visible sector probability density function of the RiemannTheta Boltzmann machine corresponds to a gaussian mixture model consisting of an infinite number of component multivariate gaussians. The weights of the mixture are given by a discrete multivariate gaussian over the hidden state space. This allows us to sample the visible sector density function in a straightforward manner. Furthermore, we show that the visible sector probability density function possesses an affine transform property, similar to the multivariate gaussian density.
 Publication:

arXiv eprints
 Pub Date:
 April 2018
 arXiv:
 arXiv:1804.07768
 Bibcode:
 2018arXiv180407768C
 Keywords:

 Statistics  Machine Learning;
 Computer Science  Machine Learning
 EPrint:
 8 pages, 6 figures