Saved in:
Bibliographic Details
Main Authors: Li, Guanxun, Smith, Aaron, Zhou, Quan
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2304.06251
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917792828096512
author Li, Guanxun
Smith, Aaron
Zhou, Quan
author_facet Li, Guanxun
Smith, Aaron
Zhou, Quan
contents We show that for any multiple-try Metropolis algorithm, one can always accept the proposal and evaluate the importance weight that is needed to correct for the bias without extra computational cost. This results in a general, convenient, and rejection-free Markov chain Monte Carlo (MCMC) sampling scheme. By further leveraging the importance sampling perspective on Metropolis--Hastings algorithms, we propose an alternative MCMC sampler on discrete spaces that is also outside the Metropolis--Hastings framework, along with a general theory on its complexity. Numerical examples suggest that the proposed algorithms are consistently more efficient than the original Metropolis--Hastings versions.
format Preprint
id arxiv_https___arxiv_org_abs_2304_06251
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Importance is Important: Generalized Markov Chain Importance Sampling Methods
Li, Guanxun
Smith, Aaron
Zhou, Quan
Computation
Methodology
Machine Learning
65C05, 60J10
We show that for any multiple-try Metropolis algorithm, one can always accept the proposal and evaluate the importance weight that is needed to correct for the bias without extra computational cost. This results in a general, convenient, and rejection-free Markov chain Monte Carlo (MCMC) sampling scheme. By further leveraging the importance sampling perspective on Metropolis--Hastings algorithms, we propose an alternative MCMC sampler on discrete spaces that is also outside the Metropolis--Hastings framework, along with a general theory on its complexity. Numerical examples suggest that the proposed algorithms are consistently more efficient than the original Metropolis--Hastings versions.
title Importance is Important: Generalized Markov Chain Importance Sampling Methods
topic Computation
Methodology
Machine Learning
65C05, 60J10
url https://arxiv.org/abs/2304.06251