The MIT PressJournals
Article navigation
Publication Cover

More About NC


Article Metrics

Altmetric

About article usage data:

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Aenean euismod bibendum laoreet. Proin gravida dolor sit amet lacus accumsan et viverra justo commodo. Proin sodales pulvinar tempor. Cum sociis natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus.


Consistency of Pseudolikelihood Estimation of Fully Visible Boltzmann Machines

Aapo Hyvärinen

Posted Online August 14, 2006
https://doi.org/10.1162/neco.2006.18.10.2283

Neural Computation

Volume 18 | Issue 10 | October 2006
p.2283-2292
Download Options

Consistency of Pseudolikelihood Estimation of Fully Visible Boltzmann Machines

Aapo Hyvärinen
https://doi.org/10.1162/neco.2006.18.10.2283
Received: September 19, 2005
Accepted: March 24, 2006
Published Online: August 14, 2006
Abstract

A Boltzmann machine is a classic model of neural computation, and a number of methods have been proposed for its estimation. Most methods are plagued by either very slow convergence or asymptotic bias in the resulting estimates. Here we consider estimation in the basic case of fully visible Boltzmann machines. We show that the old principle of pseudolikelihood estimation provides an estimator that is computationally very simple yet statistically consistent.

Aapo Hyvärinen
HIIT Basic Research Unit, Department of Computer Science, University of Helsinki, Finland
https://doi.org/10.1162/neco.2006.18.10.2283
Abstract

A Boltzmann machine is a classic model of neural computation, and a number of methods have been proposed for its estimation. Most methods are plagued by either very slow convergence or asymptotic bias in the resulting estimates. Here we consider estimation in the basic case of fully visible Boltzmann machines. We show that the old principle of pseudolikelihood estimation provides an estimator that is computationally very simple yet statistically consistent.

Aapo Hyvärinen
HIIT Basic Research Unit, Department of Computer Science, University of Helsinki, Finland
Download Options

Consistency of Pseudolikelihood Estimation of Fully Visible Boltzmann Machines

Aapo Hyvärinen
https://doi.org/10.1162/neco.2006.18.10.2283
Received: September 19, 2005
Accepted: March 24, 2006
Published Online: August 14, 2006
Abstract

A Boltzmann machine is a classic model of neural computation, and a number of methods have been proposed for its estimation. Most methods are plagued by either very slow convergence or asymptotic bias in the resulting estimates. Here we consider estimation in the basic case of fully visible Boltzmann machines. We show that the old principle of pseudolikelihood estimation provides an estimator that is computationally very simple yet statistically consistent.

Aapo Hyvärinen
HIIT Basic Research Unit, Department of Computer Science, University of Helsinki, Finland
Favorite
Track Citations
Download Citation
RSS TOC
RSS Citation