Methodological and computational aspects of parallel tempering methods in the infinite swapping limit

DOI10.1007/S10955-018-2210-YzbMATH Open1453.82106arXiv1712.06947OpenAlexW2962960075MaRDI QIDQ1731002FDOQ1731002

Authors: Eric Vanden-Eijnden, Jianfeng Lu 

Publication date: 6 March 2019

Published in: Journal of Statistical Physics (Search for Journal in Brave)

Abstract: A variant of the parallel tempering method is proposed in terms of a stochastic switching process for the coupled dynamics of replica configuration and temperature permutation. This formulation is shown to facilitate the analysis of the convergence properties of parallel tempering by large deviation theory, which indicates that the method should be operated in the infinite swapping limit to maximize sampling efficiency. The effective equation for the replica alone that arises in this infinite swapping limit simply involves replacing the original potential by a mixture potential. The analysis of the geometric properties of this potential offers a new perspective on the issues of how to choose of temperature ladder, and why many temperatures should typically be introduced to boost the sampling efficiency. It is also shown how to simulate the effective equation in this many temperature regime using multiscale integrators. Finally, similar ideas are also used to discuss extensions of the infinite swapping limits to the technique of simulated tempering.

Full work available at URL: https://arxiv.org/abs/1712.06947

Recommendations

• On the infinite swapping limit for parallel tempering
• A large deviations analysis of certain qualitative properties of parallel tempering and infinite swapping algorithms
• Analysis and optimization of certain parallel Monte Carlo methods in the low temperature limit
• Generalized parallel tempering on Bayesian inverse problems
• Ergodicity of the infinite swapping algorithm at low temperature

zbMATH Keywords

samplingparallel temperinginfinite swapping limitmultiscale integrator

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Large deviations (60F10) Parallel numerical computation (65Y05) Monte Carlo methods applied to problems in statistical mechanics (82M31)

Cites Work

• Asymptotic evaluation of certain markov process expectations for large time, I
• Nonintrusive and structure preserving multiscale integration of stiff ODEs, SDEs, and Hamiltonian systems with hidden slow dynamics via flow averaging
• Heterogeneous multiscale methods: a review
• The heterogeneous multiscale methods
• Analysis of multiscale methods for stochastic differential equations
• On the large deviation rate function for the empirical measures of reversible jump Markov processes
• On the infinite swapping limit for parallel tempering
• Numerical techniques for multi-scale dynamical systems with stochastic effects
• On HMM-like integrators and projective integration methods for systems with multiple time scales
• Feedback-optimized parallel tempering Monte Carlo
• Analysis of multiscale integrators for multiple attractors and irreversible Langevin samplers
• Dynamic variational study of chaos: spin glasses in three dimensions

Cited In (9)

• The enhanced sampling in parallel finite-time dynamics method with replica exchange
• Analysis and optimization of certain parallel Monte Carlo methods in the low temperature limit
• Ergodicity of the infinite swapping algorithm at low temperature
• A large deviations analysis of certain qualitative properties of parallel tempering and infinite swapping algorithms
• Posterior computation with the Gibbs zig-zag sampler
• Generalized parallel tempering on Bayesian inverse problems
• Monte Carlo methods for rough free energy landscapes: population annealing and parallel tempering
• State-dependent swap strategies and automatic reduction of number of temperatures in adaptive parallel tempering algorithm
• On the infinite swapping limit for parallel tempering

This page was built for publication: Methodological and computational aspects of parallel tempering methods in the infinite swapping limit

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1731002)