Generalized Parallel Tempering on Bayesian Inverse Problems

03/06/2020
by   Jonas Latz, et al.
0

In the current work we present two generalizations of the Parallel Tempering algorithm, inspired by the so-called continuous-time Infinite Swapping algorithm. Such a method, found its origins in the molecular dynamics community, and can be understood as the limit case of the continuous-time Parallel Tempering algorithm, where the (random) time between swaps of states between two parallel chains goes to zero. Thus, swapping states between chains occurs continuously. In the current work, we extend this idea to the context of time-discrete Markov chains and present two Markov chain Monte Carlo algorithms that follow the same paradigm as the continuous-time infinite swapping procedure. We analyze the convergence properties of such discrete-time algorithms in terms of their spectral gap, and implement them to sample from different target distributions. Numerical results show that the proposed methods significantly improve over more traditional sampling algorithms such as Random Walk Metropolis and (traditional) Parallel Tempering.

READ FULL TEXT

page 24

page 27

research
04/01/2019

Bisimulation for Feller-Dynkin Processes

Bisimulation is a concept that captures behavioural equivalence. It has ...
research
07/17/2017

A Discrete Bouncy Particle Sampler

Markov Chain Monte Carlo (MCMC) algorithms are statistical methods desig...
research
07/16/2018

Finding a marked node on any graph by continuous time quantum walk

We provide a new spatial search algorithm by continuous time quantum wal...
research
07/30/2015

Orthogonal parallel MCMC methods for sampling and optimization

Monte Carlo (MC) methods are widely used for Bayesian inference and opti...
research
02/14/2018

Molecular Computing for Markov Chains

In this paper, it is presented a methodology for implementing arbitraril...
research
10/14/2021

Detecting Renewal States in Chains of Variable Length via Intrinsic Bayes Factors

Markov chains with variable length are useful parsimonious stochastic mo...
research
10/30/2019

Algorithmic Randomness in Continuous-Time Markov Chains

In this paper we develop the elements of the theory of algorithmic rando...

Please sign up or login with your details

Forgot password? Click here to reset