The pseudo-marginal approach for efficient Monte Carlo computations

Abstract

We introduce a powerful and flexible MCMC algorithm for stochastic simulation. The method builds on a pseudo-marginal method originally introduced in [Genetics 164 (2003) 1139--1160], showing how algorithms which are approximations to an idealized marginal algorithm, can share the same marginal stationary distribution as the idealized method. Theoretical results are given describing the convergence properties of the proposed method, and simple numerical examples are given to illustrate the promising empirical characteristics of the technique. Interesting comparisons with a more obvious, but inexact, Monte Carlo approximation to the marginal algorithm, are also given.

Published in at http://dx.doi.org/10.1214/07-AOS574 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Cited by in corpus (185)

• A Three-Dimensional Map of Milky-Way Dust
• Markov Chain Monte Carlo Methods for Bayesian Data Analysis in Astronomy
• A Survey of Monte Carlo Methods for Parameter Estimation
• On the efficiency of pseudo-marginal random walk Metropolis algorithms
• Austerity in MCMC Land: Cutting the Metropolis-Hastings Budget
• Speeding Up MCMC by Efficient Data Subsampling
• A Survey of Stochastic Simulation and Optimization Methods in Signal Processing
• A Bayesian approach to constrained single- and multi-objective optimization
• Simultaneous reconstruction of evolutionary history and epidemiological dynamics from viral sequences with the birth-death SIR model
• The Zig-Zag Process and Super-Efficient Sampling for Bayesian Analysis of Big Data
• On Markov chain Monte Carlo methods for tall data
• Three discussions of the paper "sequential quasi-Monte Carlo sampling", by M. Gerber and N. Chopin
• Bayesian Indirect Inference Using a Parametric Auxiliary Model
• Statistical Methods and Computing for Big Data
• On Russian Roulette Estimates for Bayesian Inference with Doubly-Intractable Likelihoods
• Simulation and inference algorithms for stochastic biochemical reaction networks: from basic concepts to state-of-the-art
• GPS-ABC: Gaussian Process Surrogate Approximate Bayesian Computation
• A Review of Multiple Try MCMC algorithms for Signal Processing
• Accelerating Asymptotically Exact MCMC for Computationally Intensive Models via Local Approximations
• Particle Filters and Data Assimilation
• Piecewise Deterministic Markov Processes for Continuous-Time Monte Carlo
• Convergence properties of pseudo-marginal Markov chain Monte Carlo algorithms
• Ensemble Kalman methods for high-dimensional hierarchical dynamic space-time models
• Near-optimal protocols in complex nonequilibrium transformations
• Variational Bayes with Synthetic Likelihood
• Delayed acceptance particle MCMC for exact inference in stochastic kinetic models
• Simple simulation of diffusion bridges with application to likelihood inference for diffusions
• Flow-based sampling for fermionic lattice field theories
• Unbiased and Efficient Log-Likelihood Estimation with Inverse Binomial Sampling
• Firefly Monte Carlo: Exact MCMC with Subsets of Data
• Uniform ergodicity of the Particle Gibbs sampler
• Gauge-equivariant flow models for sampling in lattice field theories with pseudofermions
• Bayesian inference for the Brown-Resnick process, with an application to extreme low temperatures
• Speeding Up MCMC by Delayed Acceptance and Data Subsampling
• Variance bounding and geometric ergodicity of Markov chain Monte Carlo kernels for approximate Bayesian computation
• On nonnegative unbiased estimators
• Sequential Monte Carlo Methods for System Identification
• Inferring epidemiological dynamics with Bayesian coalescent inference: The merits of deterministic and stochastic models
• Coupled MCMC with a randomized acceptance probability
• Metropolized Randomized Maximum Likelihood for sampling from multimodal distributions
• Probabilistic learning of nonlinear dynamical systems using sequential Monte Carlo
• A Repelling-Attracting Metropolis Algorithm for Multimodality
• Bayesian System ID: Optimal management of parameter, model, and measurement uncertainty
• Particle Metropolis-Hastings using gradient and Hessian information
• Black-box Bayesian inference for economic agent-based models
• Gradient-free Hamiltonian Monte Carlo with Efficient Kernel Exponential Families
• Bayesian model comparison with un-normalised likelihoods
• Auxiliary Particle filtering within adaptive Metropolis-Hastings Sampling
• Forecasting: theory and practice
• Nested Sequential Monte Carlo Methods
• Perturbation Bounds for Monte Carlo within Metropolis via Restricted Approximations
• Kernel Adaptive Metropolis-Hastings
• Nonlinear System Identification: Learning while respecting physical models using a sequential Monte Carlo method
• Comparison of asymptotic variances of inhomogeneous Markov chains with application to Markov chain Monte Carlo methods
• Constructing Likelihood Functions for Interval-valued Random Variables
• Metropolis Sampling
• Importance sampling type estimators based on approximate marginal MCMC
• Accelerating pseudo-marginal Metropolis-Hastings by correlating auxiliary variables
• Pseudo-Marginal Slice Sampling
• A practical guide to pseudo-marginal methods for computational inference in systems biology
• Twisted particle filters
• Stochastic Modelling of Urban Structure
• Mode jumping MCMC for Bayesian variable selection in GLMM
• An Introduction to Twisted Particle Filters and Parameter Estimation in Non-linear State-space Models
• Bayesian inference for stochastic differential equation mixed effects models of a tumor xenography study
• Random time step probabilistic methods for uncertainty quantification in chaotic and geometric numerical integration
• Bayesian model selection for the glacial-interglacial cycle
• Efficient Continuous-Time Markov Chain Estimation
• Approximating Bayes in the 21st Century
• Bayesian Inference for Hybrid Discrete-Continuous Stochastic Kinetic Models
• Random forward models and log-likelihoods in Bayesian inverse problems
• Conditional convex orders and measurable martingale couplings
• Coupling stochastic EM and Approximate Bayesian Computation for parameter inference in state-space models
• Sequential Monte Carlo as Approximate Sampling: bounds, adaptive resampling via $\infty$-ESS, and an application to Particle Gibbs
• Gaussian variational approximation for high-dimensional state space models
• The impact of temporal sampling resolution on parameter inference for biological transport models
• Data-Free Likelihood-Informed Dimension Reduction of Bayesian Inverse Problems
• Limit theorems for sequential MCMC methods
• ABCpy: A High-Performance Computing Perspective to Approximate Bayesian Computation
• Unbiased Bayes for Big Data: Paths of Partial Posteriors
• Jump Markov Chains and Rejection-Free Metropolis Algorithms
• Multifidelity multilevel Monte Carlo to accelerate approximate Bayesian parameter inference for partially observed stochastic processes
• Sequential Monte Carlo with transformations
• Approximations of Geometrically Ergodic Reversible Markov Chains
• Learning of state-space models with highly informative observations: a tempered Sequential Monte Carlo solution
• Augmentation Schemes for Particle MCMC
• Notes to Robert et al.: Model criticism informs model choice and model comparison
• Prior normalization for certified likelihood-informed subspace detection of Bayesian inverse problems
• Generalized Parallel Tempering on Bayesian Inverse Problems
• Hamiltonian Monte Carlo with Energy Conserving Subsampling
• A comparison of Bayesian sampling algorithms for high-dimensional particle physics and cosmology applications
• A general perspective on the Metropolis-Hastings kernel
• Importance sampling correction versus standard averages of reversible MCMCs in terms of the asymptotic variance
• Accelerating Metropolis-Hastings algorithms: Delayed acceptance with prefetching
• Comparison of Markov chains via weak Poincaré inequalities with application to pseudo-marginal MCMC
• Bayesian Learning of Parameterised Quantum Circuits
• Multilevel Sequential Monte Carlo Samplers for Normalizing Constants
• Computing Bayes: From Then 'Til Now'
• Particle Metropolis adjusted Langevin algorithms for state space models
• Measuring Star-Formation Histories, Distances, and Metallicities with Pixel Color-Magnitude Diagrams I: Model Definition and Mock Tests
• Generalized Posteriors in Approximate Bayesian Computation
• Approximate Bayesian Computation by Modelling Summary Statistics in a Quasi-likelihood Framework
• Accelerating delayed-acceptance Markov chain Monte Carlo algorithms
• Generalized Bayesian Likelihood-Free Inference
• A Critical Account of Perturbation Analysis of Markov Chains
• A survey of Monte Carlo methods for noisy and costly densities with application to reinforcement learning and ABC
• Robust Bayesian Synthetic Likelihood via a Semi-Parametric Approach
• A state space approach to dynamic modeling of mouse-tracking data
• Perfect simulation using atomic regeneration with application to Sequential Monte Carlo
• Noisy Hamiltonian Monte Carlo for doubly-intractable distributions
• Introducing prior information in Weighted Likelihood Bootstrap with applications to model misspecification
• Sequentially guided MCMC proposals for synthetic likelihoods and correlated synthetic likelihoods
• Sequential Neural Posterior and Likelihood Approximation
• Spectral Subsampling MCMC for Stationary Time Series
• Lithological Tomography with the Correlated Pseudo-Marginal Method
• 3DP3: 3D Scene Perception via Probabilistic Programming
• Automatic Backward Filtering Forward Guiding for Markov processes and graphical models
• The Importance Markov Chain
• The block-Poisson estimator for optimally tuned exact subsampling MCMC
• Sequential Monte Carlo algorithms for agent-based models of disease transmission
• A numerically stable algorithm for integrating Bayesian models using Markov melding
• Estimating Reciprocal Partition Functions to Enable Design Space Sampling
• On importance-weighted autoencoders
• Simulating bridges using confluent diffusions
• Asymptotically Optimal Exact Minibatch Metropolis-Hastings
• Quantifying Age and Model Uncertainties in Paleoclimate Data and Dynamical Climate Models with a Joint Inferential Analysis
• Model comparison for Gibbs random fields using noisy reversible jump Markov chain Monte Carlo
• Likelihood-free stochastic approximation EM for inference in complex models
• Statistic Selection and MCMC for Differentially Private Bayesian Estimation
• Bayesian Spatial Monotonic Multiple Regression
• Adaptive inference over Besov spaces in the white noise model using $p$-exponential priors
• Bayesian non-parametric inference for $\Lambda$-coalescents: consistency and a parametric method
• Uncertainty quantification of a multi-component Hall thruster model at varying facility pressures
• Metropolis-Hastings with Averaged Acceptance Ratios
• Convergence of Griddy Gibbs Sampling and other perturbed Markov chains
• Tempering by Subsampling
• Modeling strength and failure variability due to porosity in additively manufactured metals
• The EAS approach to variable selection for multivariate response data in high-dimensional settings
• Anytime Monte Carlo
• Accelerating Convergence of Replica Exchange Stochastic Gradient MCMC via Variance Reduction
• Sequential Monte Carlo Methods in the nimble R Package
• Light and Widely Applicable MCMC: Approximate Bayesian Inference for Large Datasets
• Mini-batch Metropolis-Hastings MCMC with Reversible SGLD Proposal
• Parallelising MCMC via Random Forests
• Parameter inference method for stochastic single-cell dynamics from tree-structured data
• Synthetic likelihood in misspecified models
• Evaluating and Optimizing Network Sampling Designs: Decision Theory and Information Theory Perspectives
• Efficient Bernoulli factory MCMC for intractable posteriors
• A determinant-free method to simulate the parameters of large Gaussian fields
• Pseudo-extended Markov chain Monte Carlo
• Quasi-Newton particle Metropolis-Hastings
• Approximate Bayesian inference from noisy likelihoods with Gaussian process emulated MCMC
• Rao-Blackwellization in the MCMC era
• A Multi-armed Bandit MCMC, with applications in sampling from doubly intractable posterior
• Scalable Approximate Inference and Some Applications
• Revisiting the balance heuristic for estimating normalising constants
• f-SAEM: A fast Stochastic Approximation of the EM algorithm for nonlinear mixed effects models
• Accelerating Metropolis-within-Gibbs sampler with localized computations of differential equations
• Likelihood-free approximate Gibbs sampling
• Efficient posterior sampling for high-dimensional imbalanced logistic regression
• Simulation-based inference for stochastic nonlinear mixed-effects models with applications in systems biology
• Neural Bayes estimation and selection for complex bivariate extremal dependence models
• The Normal-Generalised Gamma-Pareto process: A novel pure-jump Lévy process with flexible tail and jump-activity properties
• The Feature-First Block Model
• Towards Data-Conditional Simulation for ABC Inference in Stochastic Differential Equations
• A Drift Homotopy Implicit Particle Filter Method for Nonlinear Filtering problems
• A High-Dimensional Particle Filter Algorithm
• POPE: Post Optimization Posterior Evaluation of Likelihood Free Models
• ABC-Di: Approximate Bayesian Computation for Discrete Data
• Scalable couplings for the random walk Metropolis algorithm
• Bayesian Inference of Reproduction Number from Epidemiological and Genetic Data Using Particle MCMC
• Perturbation Analysis of Markov Chain Monte Carlo for Graphical Models
• On integral priors for multiple comparison in Bayesian model selection
• Bring the noise: exact inference from noisy simulations in collider physics
• Optimality in Noisy Importance Sampling
• Gaussian Processes to speed up MCMC with automatic exploratory-exploitation effect
• Probabilistic Programs with Stochastic Conditioning
• Exact Convergence Rate Analysis of the Independent Metropolis-Hastings Algorithms
• Sampling Methods for Bayesian Inference Involving Convergent Noisy Approximations of Forward Maps
• A Bayesian framework for case-cohort Cox regression: application to dietary epidemiology
• Stability of Conditional Sequential Monte Carlo
• On MCMC for variationally sparse Gaussian processes: A pseudo-marginal approach
• The Node-wise Pseudo-marginal Method
• A Point Mass Proposal Method for Bayesian State-Space Model Fitting
• Estimation and uncertainty quantification for the output from quantum simulators