Optimal Scaling Results for Moreau-Yosida Metropolis-adjusted Langevin Algorithms

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Crucinio, Francesca R., Durmus, Alain, Jiménez, Pablo, Roberts, Gareth O.
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911925307179008
author Crucinio, Francesca R.
Durmus, Alain
Jiménez, Pablo
Roberts, Gareth O.
author_facet Crucinio, Francesca R.
Durmus, Alain
Jiménez, Pablo
Roberts, Gareth O.
contents We consider a recently proposed class of MCMC methods which uses proximity maps instead of gradients to build proposal mechanisms which can be employed for both differentiable and non-differentiable targets. These methods have been shown to be stable for a wide class of targets, making them a valuable alternative to Metropolis-adjusted Langevin algorithms (MALA); and have found wide application in imaging contexts. The wider stability properties are obtained by building the Moreau-Yosida envelope for the target of interest, which depends on a parameter $λ$. In this work, we investigate the optimal scaling problem for this class of algorithms, which encompasses MALA, and provide practical guidelines for the implementation of these methods.
format Preprint
id arxiv_https___arxiv_org_abs_2301_02446
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Optimal Scaling Results for Moreau-Yosida Metropolis-adjusted Langevin Algorithms
Crucinio, Francesca R.
Durmus, Alain
Jiménez, Pablo
Roberts, Gareth O.
Computation
Probability
Statistics Theory
65C05, 60F05
We consider a recently proposed class of MCMC methods which uses proximity maps instead of gradients to build proposal mechanisms which can be employed for both differentiable and non-differentiable targets. These methods have been shown to be stable for a wide class of targets, making them a valuable alternative to Metropolis-adjusted Langevin algorithms (MALA); and have found wide application in imaging contexts. The wider stability properties are obtained by building the Moreau-Yosida envelope for the target of interest, which depends on a parameter $λ$. In this work, we investigate the optimal scaling problem for this class of algorithms, which encompasses MALA, and provide practical guidelines for the implementation of these methods.
title Optimal Scaling Results for Moreau-Yosida Metropolis-adjusted Langevin Algorithms
topic Computation
Probability
Statistics Theory
65C05, 60F05
url https://arxiv.org/abs/2301.02446