A Family of MCMC Methods on Implicitly Defined Manifolds

Marcus Brubaker, Mathieu Salzmann, Raquel Urtasun ; JMLR W&CP 22: 161-172, 2012.

Abstract

Traditional MCMC methods are only applicable to distributions which can be defined on $\mathbb{R}^n$. However, there exist many application domains where the distributions cannot easily be defined on a Euclidean space. To address this limitation, we propose a general constrained version of Hamiltonian Monte Carlo, and give conditions under which the Markov chain is convergent. Based on this general framework we define a family of MCMC methods which can be applied to sample from distributions on non-linear manifolds. We demonstrate the effectiveness of our approach on a variety of problems including sampling from the Bingham-von Mises-Fisher distribution, collaborative filtering and human pose estimation.




Home Page

Papers

Submissions

News

Scope

Editorial Board

Announcements

Proceedings

Open Source Software

Search

Login



RSS Feed

Page last modified on Thu April 26 2012 13:56 2012.

Copyright @ JMLR 2012. All rights reserved.