Publica
Hier finden Sie wissenschaftliche Publikationen aus den FraunhoferInstituten. Evaluating the RBM without integration using PDF projection
 European Association for Signal Processing EURASIP: 25th European Signal Processing Conference, EUSIPCO 2017 : 27 August  2 September 2017, Kos Island, Greece Kos, 2017 ISBN: 9780992862671 pp.858862 
 European Signal Processing Conference (EUSIPCO) <25, 2017, Kos> 

 English 
 Conference Paper, Electronic Publication 
 Fraunhofer FKIE () 
Abstract
In this paper, we apply probability density function (PDF) projection to arrive at an exact closedform expression for the marginal distribution of the visible data of a restricted Boltzmann machine (RBM) without requiring integrating over the distribution of the hidden variables or needing to know the partition function. We express the visible data marginal as a projected PDF based on a set of sufficient statistics. When a Gaussian mixture model (GMM) is used to estimate the PDF of the sufficient statistics, then we arrive at a combined RBM/GMM model that serves as a generalpurpose PDF estimator and Bayesian classifier. The approach extends recusively to compute the input distribution of a multilayer network. We demonstrate the method using a reduced subset of the MNIST handwritten character data set.