Langevin diffusions and the Metropolis-adjusted Langevin algorithm

被引：71

作者：

Xifara, T. ^{[1
,2
]}

Sherlock, C. ^{[2
]}

Livingstone, S. ^{[3
]}

Byrne, S. ^{[3
]}

Girolami, M. ^{[3
]}

机构：

[1] Univ Calif Santa Cruz, Dept Appl Math & Stat, Santa Cruz, CA 95064 USA

[2] Univ Lancaster, Dept Math & Stat, Lancaster LA1 4YW, England

[3] UCL, Dept Stat Sci, London WC1E 6BT, England

来源：

STATISTICS & PROBABILITY LETTERS | 2014年 / 91卷

基金：

英国工程与自然科学研究理事会;

关键词：

Diffusions; Markov chain Monte Carlo; Metropolis-adjusted Langevin algorithm; Riemannian manifolds;

D O I：

10.1016/j.spl.2014.04.002

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

We describe a Langevin diffusion with a target stationary density with respect to Lebesgue measure, as opposed to the volume measure of a previously-proposed diffusion. The two are sometimes equivalent but in general distinct and lead to different Metropolis-adjusted Langevin algorithms, which we compare. (C) 2014 Elsevier B.V. All rights reserved.

引用

页码：14 / 19

页数：6