Moving Target Monte Carlo
Abstract
The Markov Chain Monte Carlo (MCMC) methods are popular when considering sampling from a high-dimensional random variable with possibly unnormalised probability density and observed data . However, MCMC requires evaluating the posterior distribution of the proposed candidate at each iteration when constructing the acceptance rate. This is costly when such evaluations are intractable. In this paper, we introduce a new non-Markovian sampling algorithm called Moving Target Monte Carlo (MTMC). The acceptance rate at -th iteration is constructed using an iteratively updated approximation of the posterior distribution instead of . The true value of the posterior is only calculated if the candidate is accepted. The approximation utilises these evaluations and converges to as . A proof of convergence and estimation of convergence rate in different situations are given.
Cite
@article{arxiv.2003.04873,
title = {Moving Target Monte Carlo},
author = {Haoyun Ying and Keheng Mao and Klaus Mosegaard},
journal= {arXiv preprint arXiv:2003.04873},
year = {2020}
}