Asymptotically Independent Markov Sampling: a new MCMC scheme for Bayesian Inference

  title={Asymptotically Independent Markov Sampling: a new MCMC scheme for Bayesian Inference},
  author={James L. Beck and Konstantin Zuev},
  journal={arXiv: Computation},
  • J. Beck, K. Zuev
  • Published 9 October 2011
  • Computer Science
  • arXiv: Computation
In Bayesian statistics, many problems can be expressed as the evaluation of the expectation of a quantity of interest with respect to the posterior distribution. Standard Monte Carlo method is often not applicable because the encountered posterior distributions cannot be sampled directly. In this case, the most popular strategies are the importance sampling method, Markov chain Monte Carlo, and annealing. In this paper, we introduce a new scheme for Bayesian inference, called Asymptotically… 
Transitional annealed adaptive slice sampling for Gaussian process hyper-parameter estimation
A more complete Bayesian inference is carried out, that combines Slice Sampling with some recently developed Sequential Monte Carlo samplers, that improves the mixing in the sampling through delayed-rejection, the inclusion of an annealing scheme akin to Asymptotically Independent Markov Sampling and parallelisation via Transitional Markov Chain Monte Carlo.
Global optimization for performance-based design using the Asymptotically Independent Markov Sampling Method
A new efficient stochastic simulation method, AIMS-OPT, for approximating the set of globally optimal solutions when solving optimization problems such as optimal performance-based design problems, based on Asymptotically Independent Markov Sampling.
Stochastic Spectral Methods for Linear Bayesian Inference
In this work, a family of linear Bayesian approaches is presented which directly results in a representation of the posterior and a specific implementation is discussed which overcomes some of the difficulties that remained unsolved in related approaches.
Stochastic Methods for Emulation, Calibration and Reliability Analysis of Engineering Models
New MCMC algorithms are presented which allow the use of full Bayesian emulators by sampling from their respective multimodal posteriors, and the GP surrogate model’s probabilistic statements are exploited and the data assimilation process is improved.
History matching with probabilistic emulators and active learning
A Gaussian process provides a complete probabilistic output that is incorporated into the implausibility measure and the identification of regions of interest is accomplished with recently developed annealing sampling techniques.
Particle filtering and marginalization for parameter identification in structural systems
In structural health monitoring, one wishes to use available measurements from a structure to assess structural condition, localize damage if present, and quantify remaining life. Nonlinear system
Bayesian inversion for finite fault earthquake source models I—theory and algorithm
The estimation of finite fault earthquake source models is an inherently underdetermined problem: there is no unique solution to the inverse problem of determining the rupture history at depth as a
A Sequential Algorithm for Signal Segmentation
This work analyzes 15-min samples of an acoustic signal, and applies a sequential algorithm with the only assumption that an event alters the energy of the signal, entirely based on Bayesian methods.
Bayesian-based Finite Element Model Updating, Damage Detection, and Uncertainty Quantification for Cable-stayed Bridges
This research seeks to address some of the key challenges in SHM of large-scale civil infrastructure such as analyzing a huge quantity of measured data for system identification, dealing with uncertainty in measurements and analytical models of structures, performing a real-world application of Bayesian Finite Element (FE) model updating, and Bayesian-based damage detection.


Asymptotically Independent Markov Sampling: A New Markov Chain Monte Carlo Scheme for Bayesian Interference
A new scheme for Bayesian inference, called Asymptotically Independent Markov Sampling (AIMS), which is based on the above methods is introduced and it is shown that, under certain conditions, the AIMS algorithm produces a uniformly ergodic Markov chain.
Annealing Markov chain Monte Carlo with applications to ancestral inference
This work proposes MCMC methods distantly related to simulated annealing, which simulate realizations from a sequence of distributions, allowing the distribution being simulated to vary randomly over time.
Monte Carlo Sampling Methods Using Markov Chains and Their Applications
SUMMARY A generalization of the sampling method introduced by Metropolis et al. (1953) is presented along with an exposition of the relevant theory, techniques of application and methods and
Transitional Markov Chain Monte Carlo Method for Bayesian Model Updating, Model Class Selection, and Model Averaging
This paper presents a newly developed simulation-based approach for Bayesian model updating, model class selection, and model averaging called the transitional Markov chain Monte Carlo (TMCMC) approach, motivated by the adaptive Metropolis–Hastings method.
Markov Chains for Exploring Posterior Distributions
Several Markov chain methods are available for sampling from a posterior distribution. Two important examples are the Gibbs sampler and the Metropolis algorithm. In addition, several strategies are
Probabilistic Inference Using Markov Chain Monte Carlo Methods
The role of probabilistic inference in artificial intelligence is outlined, the theory of Markov chains is presented, and various Markov chain Monte Carlo algorithms are described, along with a number of supporting techniques.
Bayesian Inference in Econometric Models Using Monte Carlo Integration
Methods for the systematic application of Monte Carlo integration with importance sampling to Bayesian inference are developed. Conditions under which the numerical approximation converges almost
Annealed importance sampling
It is shown how one can use the Markov chain transitions for such an annealing sequence to define an importance sampler, which can be seen as a generalization of a recently-proposed variant of sequential importance sampling.
Metropolized independent sampling with comparisons to rejection sampling and importance sampling
In this paper, a special Metropolis-Hastings algorithm, Metropolized independent sampling, proposed first in Hastings (1970), is studied in full detail and shown to be superior to rejection sampling in two respects: asymptotic efficiency and ease of computation.
Markov Chain Monte Carlo conver-gence diagnostics: a comparative review
All of the methods in this work can fail to detect the sorts of convergence failure that they were designed to identify, so a combination of strategies aimed at evaluating and accelerating MCMC sampler convergence are recommended.