Annealed Importance Sampling

Abstract

Simulated annealing - moving from a tractable distribution to a distribution of interest via a sequence of intermediate distributions - has traditionally been used as an inexact method of handling isolated modes in Markov chain samplers. Here, it is shown how one can use the Markov chain transitions for such an annealing sequence to define an importance sampler. The Markov chain aspect allows this method to perform acceptably even for high-dimensional problems, where finding good importance sampling distributions would otherwise be very difficult, while the use of importance weights ensures that the estimates found converge to the correct values as the number of annealing runs increases. This annealed importance sampling procedure resembles the second half of the previously-studied tempered transitions, and can be seen as a generalization of a recently-proposed variant of sequential importance sampling. It is also related to thermodynamic integration methods for estimating ratios of normalizing constants. Annealed importance sampling is most attractive when isolated modes are present, or when estimates of normalizing constants are required, but it may also be more generally useful, since its independent sampling allows one to bypass some of the problems of assessing convergence and autocorrelation in Markov chain samplers.

Cited by in corpus (25)

• MCMC using Hamiltonian dynamics
• Fighting Sample Degeneracy and Impoverishment in Particle Filters: A Review of Intelligent Approaches
• A Quantum Fluctuation Theorem
• Properties of Nested Sampling
• Microscopic analysis of Clausius-Duhem processes
• On Russian Roulette Estimates for Bayesian Inference with Doubly-Intractable Likelihoods
• On the use of marginal posteriors in marginal likelihood estimation via importance-sampling
• Bayesian Evidence and Model Selection
• Learning Feature Hierarchies with Centered Deep Boltzmann Machines
• EDropout: Energy-Based Dropout and Pruning of Deep Neural Networks
• Bayesian inversion in resin transfer molding
• Modified Hamiltonian Monte Carlo for Bayesian inference
• Multi-sensor Joint Adaptive Birth Sampler for Labeled Random Finite Set Tracking
• Divide-and-Conquer with Sequential Monte Carlo
• A Mixture-Based Approach to Regional Adaptation for MCMC
• Free energy calculation of crystalline solids using normalizing flow
• MCMC-driven importance samplers
• Inferring effective couplings with Restricted Boltzmann Machines
• Adaptive sequential Monte Carlo for posterior inference and model selection among complex geological priors
• Bayes Model Selection with Path Sampling: Factor Models and Other Examples
• Transitional annealed adaptive slice sampling for Gaussian process hyper-parameter estimation
• Extending fragment-based free energy calculations with library Monte Carlo simulation: Annealing in interaction space
• Multidimensional integration through Markovian sampling under steered function morphing: a physical guise from statistical mechanics
• A connection between Tempering and Entropic Mirror Descent
• Stochastic Gradient Annealed Importance Sampling for Efficient Online Marginal Likelihood Estimation