An adaptive sequential Monte Carlo sampler

DOI10.1214/13-BA814MaRDI QIDQ908007zbMATH OpenOpenAlexFDO

Authors Paul Fearnhead, Benjamin M. Taylor

Publication date 2 February 2016

Published in Bayesian Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1005.1193, https://projecteuclid.org/euclid.ba/1369407558

zbMATH Keywords

adaptive MCMCadaptive sequential Monte CarloBayesian mixture analysisoptimal scalingstochastic optimization

Mathematics Subject Classification ID

Computational methods in Markov chains (60J22) Sampling theory, sample surveys (62D05) Bayesian inference (62F15) Monte Carlo methods (65C05)

Abstract: Sequential Monte Carlo (SMC) methods are not only a popular tool in the analysis of state space models, but offer an alternative to MCMC in situations where Bayesian inference must proceed via simulation. This paper introduces a new SMC method that uses adaptive MCMC kernels for particle dynamics. The proposed algorithm features an online stochastic optimization procedure to select the best MCMC kernel and simultaneously learn optimal tuning parameters. Theoretical results are presented that justify the approach and give guidance on how it should be implemented. Empirical results, based on analysing data from mixture models, show that the new adaptive SMC algorithm (ASMC) can both choose the best MCMC kernel, and learn an appropriate scaling for it. ASMC with a choice between kernels outperformed the adaptive MCMC algorithm of Haario et al. (1998) in 5 out of the 6 cases considered.

Recommendations

• On the convergence of adaptive sequential Monte Carlo methods
• On adaptive resampling strategies for sequential Monte Carlo methods
• Sequential Monte Carlo Samplers
• scientific article; zbMATH DE number 5007617
• \(\mathrm{SMC}^2\): an efficient algorithm for sequential analysis of state space models

Cited in

(43)

• Interacting sequential Monte Carlo samplers for trans-dimensional simulation
• Sequential Monte Carlo EM for multivariate probit models
• Likelihood-free Bayesian estimation of multivariate quantile distributions
• Gaussian process hyper-parameter estimation using parallel asymptotically independent Markov sampling
• Transdimensional sequential Monte Carlo using variational Bayes -- SMCVB
• Sequentially adaptive Bayesian learning algorithms for inference and optimization
• A quasi-Bayesian perspective to online clustering
• Adaptive stratified Monte Carlo algorithm for numerical computation of integrals
• Accelerating sequential Monte Carlo with surrogate likelihoods
• Ensemble Kalman filter based sequential Monte Carlo sampler for sequential Bayesian inference
• Efficient real-time monitoring of an emerging influenza pandemic: how feasible?
• Subsampling sequential Monte Carlo for static Bayesian models
• Variance estimation in adaptive sequential Monte Carlo
• Sequential Monte Carlo with transformations
• Sequential Monte Carlo samplers with independent Markov chain Monte Carlo proposals
• Sequential Monte Carlo for Bayesian sequentially designed experiments for discrete data
• Layered adaptive importance sampling
• Adaptive particle allocation in iterated sequential Monte Carlo via approximating meta-models
• A pseudo-marginal sequential Monte Carlo online smoothing algorithm
• On the convergence of adaptive sequential Monte Carlo methods
• Antithetic sampling for sequential Monte Carlo methods with application to state-space models
• Particle Markov chain Monte Carlo for efficient numerical simulation
• Sequential Monte Carlo Samplers
• On adaptive resampling strategies for sequential Monte Carlo methods
• Efficient Sequential Monte-Carlo Samplers for Bayesian Inference
• Adapting the Number of Particles in Sequential Monte Carlo Methods Through an Online Scheme for Convergence Assessment
• A semiautomatic method for history matching using sequential Monte Carlo
• Bayesian model comparison with the Hyvärinen score: computation and consistency
• Combined use of importance weights and resampling weights in sequential Monte Carlo methods
• Alive SMC\(^{2}\): Bayesian model selection for low-count time series models with intractable likelihoods
• An Invitation to Sequential Monte Carlo Samplers
• Finite sample complexity of sequential Monte Carlo estimators on multimodal target distributions
• Automatically adapting the number of state particles in \(\text{SMC}^2\)
• Adaptive tuning of Hamiltonian Monte Carlo within sequential Monte Carlo
• Sequentially interacting Markov chain Monte Carlo methods
• Unbiased and consistent nested sampling via sequential Monte Carlo
• On sequential Monte Carlo, partial rejection control and approximate Bayesian computation
• Online semiparametric regression via sequential Monte Carlo
• A Bayesian inverse approach to proton therapy dose delivery verification
• Nested ensemble Kalman filter for static parameter inference in nonlinear state-space models
• Sequential capture-recapture method
• Sequential Monte Carlo on large binary sampling spaces
• Adaptive sequential Monte Carlo by means of mixture of experts

This page was built for publication: An adaptive sequential Monte Carlo sampler

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q908007)