Repository logo

Infoscience

  • English
  • French
Log In
Logo EPFL, École polytechnique fédérale de Lausanne

Infoscience

  • English
  • French
Log In
  1. Home
  2. Academic and Research Output
  3. Journal articles
  4. Generalized parallel tempering on Bayesian inverse problems
 
research article

Generalized parallel tempering on Bayesian inverse problems

Latz, Jonas
•
Madrigal-Cianci, Juan P.
•
Nobile, Fabio  
Show more
September 1, 2021
Statistics And Computing

In the current work we present two generalizations of the Parallel Tempering algorithm in the context of discrete-timeMarkov chainMonteCarlo methods for Bayesian inverse problems. These generalizations use state-dependent swapping rates, inspired by the so-called continuous time Infinite Swapping algorithm presented in Plattner et al. (J Chem Phys 135(13):134111, 2011). We analyze the reversibility and ergodicity properties of our generalized PT algorithms. Numerical results on sampling from different target distributions, show that the proposed methods significantly improve sampling efficiency over more traditional sampling algorithms such as Random Walk Metropolis, preconditioned Crank-Nicolson, and (standard) Parallel Tempering.

  • Files
  • Details
  • Metrics
Type
research article
DOI
10.1007/s11222-021-10042-6
Web of Science ID

WOS:000692407900001

Author(s)
Latz, Jonas
Madrigal-Cianci, Juan P.
Nobile, Fabio  
Tempone, Raul
Date Issued

2021-09-01

Published in
Statistics And Computing
Volume

31

Issue

5

Start page

67

Subjects

Computer Science, Theory & Methods

•

Statistics & Probability

•

Computer Science

•

Mathematics

•

bayesian inversion

•

parallel tempering

•

infinite swapping

•

markov chain monte carlo

•

uncertainty quantification

•

monte-carlo methods

•

mcmc methods

•

convergence

•

algorithm

Editorial or Peer reviewed

REVIEWED

Written at

EPFL

EPFL units
CSQI  
RelationURL/DOI

IsNewVersionOf

https://infoscience.epfl.ch/record/276213
Available on Infoscience
September 25, 2021
Use this identifier to reference this record
https://infoscience.epfl.ch/handle/20.500.14299/181703
Logo EPFL, École polytechnique fédérale de Lausanne
  • Contact
  • infoscience@epfl.ch

  • Follow us on Facebook
  • Follow us on Instagram
  • Follow us on LinkedIn
  • Follow us on X
  • Follow us on Youtube
AccessibilityLegal noticePrivacy policyCookie settingsEnd User AgreementGet helpFeedback

Infoscience is a service managed and provided by the Library and IT Services of EPFL. © EPFL, tous droits réservés