On adaptive resampling strategies for sequential Monte Carlo methods

Abstract

Sequential Monte Carlo (SMC) methods are a class of techniques to sample approximately from any sequence of probability distributions using a combination of importance sampling and resampling steps. This paper is concerned with the convergence analysis of a class of SMC methods where the times at which resampling occurs are computed online using criteria such as the effective sample size. This is a popular approach amongst practitioners but there are very few convergence results available for these methods. By combining semigroup techniques with an original coupling argument, we obtain functional central limit theorems and uniform exponential concentration estimates for these algorithms.

Published in at http://dx.doi.org/10.3150/10-BEJ335 the Bernoulli (http://isi.cbs.nl/bernoulli/) by the International Statistical Institute/Bernoulli Society (http://isi.cbs.nl/BS/bshome.htm)

Cited by in corpus (41)

• Effective Sample Size for Importance Sampling based on discrepancy measures
• Multilevel Sequential Monte Carlo Samplers
• Improving Population Monte Carlo: Alternative Weighting and Resampling Schemes
• Monte Carlo Confidence Sets for Identified Sets
• On the role of interaction in sequential Monte Carlo algorithms
• Stability properties of some particle filters
• Implicit particle methods and their connection with variational data assimilation
• Large deviations of Lyapunov exponents
• Rare event computation in deterministic chaotic systems using genealogical particle analysis
• Divide-and-Conquer with Sequential Monte Carlo
• Particle Smoothing for Hidden Diffusion Processes: Adaptive Path Integral Smoother
• A Stable Particle Filter in High-Dimensions
• Linear Variance Bounds for Particle Approximations of Time-Homogeneous Feynman-Kac Formulae
• Optimized Population Monte Carlo
• Marginalizing Gaussian Process Hyperparameters using Sequential Monte Carlo
• A coherent structure approach for parameter estimation in Lagrangian Data Assimilation
• Approximate Smoothing and Parameter Estimation in High-Dimensional State-Space Models
• Sequential Monte Carlo as Approximate Sampling: bounds, adaptive resampling via $\infty$-ESS, and an application to Particle Gibbs
• On the Stability of Sequential Monte Carlo Methods in High Dimensions
• Error Bounds and Normalizing Constants for Sequential Monte Carlo in High Dimensions
• Nonasymptotic analysis of adaptive and annealed Feynman-Kac particle models
• Central Limit Theorems for Coupled Particle Filters
• How biased is your model? Concentration Inequalities, Information and Model Bias
• Fast and numerically stable particle-based online additive smoothing: the AdaSmooth algorithm
• Efficient learning in ABC algorithms
• On the Convergence of Adaptive Sequential Monte Carlo Methods
• Marginal sequential Monte Carlo for doubly intractable models
• Annealed Flow Transport Monte Carlo
• Transitional annealed adaptive slice sampling for Gaussian process hyper-parameter estimation
• Bayesian Inference for Logistic Regression Models Using Sequential Posterior Simulation
• Quantitative approximations of evolving probability measures and sequential Markov Chain Monte Carlo methods
• Sequential Kalman Tuning of the $t$-preconditioned Crank-Nicolson algorithm: efficient, adaptive and gradient-free inference for Bayesian inverse problems
• Bayesian Sparsity-Path-Analysis of Genetic Association Signal using Generalized t Priors
• Sequential Monte Carlo Methods for Option Pricing
• A critical analysis of resampling strategies for the regularized particle filter
• Dynamic Weight Importance Sampling for Low Cost Spatiotemporal Sensing
• Computational Methods for a Class of Network Models
• Fluctuations, stability and instability of a distributed particle filter with local exchange
• A Bayesian Mixture of Lasso Regressions with t-Errors
• Adaptive sequential Monte Carlo by means of mixture of experts
• Feynman-Kac particle integration with geometric interacting jumps