Convergence of adaptive mixtures of importance sampling schemes
From MaRDI portal
Publication:997389
DOI10.1214/009053606000001154zbMATH Open1132.60022arXiv0708.0711OpenAlexW2996433517WikidataQ60461503 ScholiaQ60461503MaRDI QIDQ997389FDOQ997389
Authors: Randal Douc, Arnaud Guillin, Jean-Michel Marin, Christian P. Robert Robert 
Publication date: 23 July 2007
Published in: The Annals of Statistics (Search for Journal in Brave)
Abstract: In the design of efficient simulation algorithms, one is often beset with a poor choice of proposal distributions. Although the performance of a given simulation kernel can clarify a posteriori how adequate this kernel is for the problem at hand, a permanent on-line modification of kernels causes concerns about the validity of the resulting algorithm. While the issue is most often intractable for MCMC algorithms, the equivalent version for importance sampling algorithms can be validated quite precisely. We derive sufficient convergence conditions for adaptive mixtures of population Monte Carlo algorithms and show that Rao--Blackwellized versions asymptotically achieve an optimum in terms of a Kullback divergence criterion, while more rudimentary versions do not benefit from repeated updating.
Full work available at URL: https://arxiv.org/abs/0708.0711
Recommendations
- Adaptive mixture importance sampling
- Convergence rates for optimised adaptive importance samplers
- Convergence and efficiency of adaptive importance sampling techniques with partial biasing
- Importance sampling schemes for evidence approximation in mixture models
- Exponential convergence of adaptive importance sampling for Markov chains
- Efficient importance sampling in mixture frameworks
- Safe adaptive importance sampling: a mixture approach
- Stochastic adaptation of importance sampler
Numerical analysis or methods applied to Markov chains (65C40) Central limit and other weak theorems (60F05)
Cites Work
- Sequential Monte Carlo Methods in Practice
- Sequential Monte Carlo Samplers
- Weak convergence and optimal scaling of random walk Metropolis algorithms
- Markov chains for exploring posterior distributions. (With discussion)
- Title not available (Why is that?)
- Title not available (Why is that?)
- An adaptive Metropolis algorithm
- Central limit theorem for sequential Monte Carlo methods and its application to Bayesian inference
- Inference in hidden Markov models.
- Adaptive proposal distribution for random walk Metropolis algorithm
- Self-regenerative Markov chain Monte Carlo with adaptation
- Rates of convergence of the Hastings and Metropolis algorithms
- Adaptive Markov Chain Monte Carlo through Regeneration
- Weighted Average Importance Sampling and Defensive Mixture Distributions
- Intrinsic losses
- Recursive Monte Carlo filters: algorithms and theoretical analysis
- Title not available (Why is that?)
- Title not available (Why is that?)
- Iterated importance sampling in missing data problems
- Minimum variance importance samplingviaPopulation Monte Carlo
- Limit theorems for weighted samples with applications to sequential Monte Carlo methods
Cited In (37)
- Bayesian model averaging in astrophysics: a review
- Layered adaptive importance sampling
- Collective proposal distributions for nonlinear MCMC samplers: mean-field theory and fast implementation
- On variance stabilisation in population Monte Carlo by double Rao-Blackwellisation
- Multifidelity importance sampling
- Minimum variance importance samplingviaPopulation Monte Carlo
- A tutorial on approximate Bayesian computation
- Generalized multiple importance sampling
- Incremental Mixture Importance Sampling With Shotgun Optimization
- Sequential Monte Carlo with transformations
- Iterative importance sampling algorithms for parameter estimation
- Gradient-based adaptive importance samplers
- Use in practice of importance sampling for repeated MCMC for Poisson models
- Convergence rates for optimised adaptive importance samplers
- Likelihood free inference for Markov processes: a comparison
- Iterative Bayesian inversion with Gaussian mixtures: finite sample implementation and large sample asymptotics
- Adaptive multiple importance sampling
- Adaptive multiple importance sampling for Gaussian processes
- Combining multiple surrogate models to accelerate failure probability estimation with expensive high-fidelity models
- Deep Importance Sampling Using Tensor Trains with Application to a Priori and a Posteriori Rare Events
- Target-aware Bayesian inference: how to beat optimal conventional estimators
- Safe adaptive importance sampling: a mixture approach
- Infinite-dimensional gradient-based descent for alpha-divergence minimisation
- Method for approximating target distribution of importance sampling
- Consistency of adaptive importance sampling and recycling schemes
- Parallelizing MCMC sampling via space partitioning
- Convergence of Monte Carlo distribution estimates from rival samplers
- Population Monte Carlo algorithm in high dimensions
- Efficient importance sampling in mixture frameworks
- Ensemble transport adaptive importance sampling
- Transport map accelerated adaptive importance sampling, and application to inverse problems arising from multiscale stochastic reaction networks
- Stochastic adaptation of importance sampler
- A survey of sequential Monte Carlo methods for economics and finance
- Approximate Bayesian computational methods
- Particle methods for statistical inference and design optimization
- On convergence of properly weighted samples to the target distribution
- Accelerating MCMC algorithms
This page was built for publication: Convergence of adaptive mixtures of importance sampling schemes
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q997389)
