Adaptive Rejection Sampling for Gibbs Sampling

By By W. R. GILKSt. 2010

Subfields: Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Probabilistic & Bayesian Methods

Citations in the corpus

Generalizing Hamiltonian Monte Carlo with Neural Networks (2017, background): “In addition, one could draw inspiration from the adaptive literature (Haario et al., 2001; Andrieu & Thoms, 2008) or component-wise strategies (Gilks & Wild, 1992).”
A* Sampling (2014, methodology): “Some are specialized to specific distributions [5], but exact generic methods are based either on (adaptive) rejection sampling [6, 7, 8] or Markov Chain Monte Carlo (MCMC) methods where convergence to the stationary distribution can be guaranteed [9, 10, 11].”
A Bayesian model for repeated measures zero-inflated count data with application to outpatient psychiatric service use (2010, background): “The full conditionals for the above models do not have closed forms; however, since the likelihoods are log-concave, adaptive rejection sampling or adaptive rejection Metropolis sampling can be used (Gilks and Wild, 1992; Gilks et al., 1995).”
Regression and Classification Using Gaussian Process Priors (2009, methodology): “The final conditional density is log-concave, and hence can be sampled from efficiently using the adaptive rejection method of Gilks and Wild (1992).”
Beta-Negative Binomial Process and Poisson Factor Analysis (2011, methodology): “Note that the adaptive rejection sampling in Gilks and Wild (1992) may also be used to sample r k .”
Log-Concavity and Strong Log-Concavity: a review (2014, methodology): “In Hargé [1999], the author gives an elegant partial answer to Problem (9.21), using semigroup techniques. The Gaussian correlation conjecture has indeed been proved to hold when d = 2 by Pitt [1977] and by Hargé [1999] when one of the sets is a symmetric ellipsoid and the other is convex symmetric.”
Conditional (2019, unknown): “To effectively accomplish face de-identification, we meticulously integrate multidimensional conditions to ensure the protection of individual identities while navigating the nuances of privacy preservation.”
Weighted exponential distribution: properties and different methods of estimation (2015, methodology): “[2] Since, p(β|α, x) has log-concave density and p(α|β, x) has a finite maximum point thus using the idea of Gilks and Wild,[19] adaptive acceptance–rejection sampling, it is possible to generate data from p(β|α, x) and p(α|β, x), respectively.”
A Unified Framework for Fitting Bayesian Semiparametric Models to Arbitrarily Censored Survival Data, Including Spatially Referenced Data (2017, methodology): “Furthermore, they update one regression coefficient β j at a time for j = 1, . . . , p via the ARS method (Gilks and Wild 1992), and it is well-known that any component-at-a-time sampler will suffer from poor mixing if there is strong correlation in the posterior.”
Missing data methods in longitudinal studies: a review (2009, methodology): “If this distribution is from the class of concave log-densities, then Gibbs sampling from (34) is straightforward using the adaptive rejection algorithm of Gilks and Wild (1992).”
Bayesian analysis for step-stress accelerated life testing using weibull proportional hazard model (2014, methodology): “See the algorithm details in Gilks and Wild (1992).”
MCMC ESTIMATION OF LÉVY JUMP MODELS USING STOCK AND OPTION PRICES (2011, methodology): “So it is very efficient to use the ARS algorithm of Gilks (1992) to sample from this posterior distribution.”
The OS* Algorithm: a Joint Approach to Exact Optimization and Sampling (2012, result): “In order to do that, some authors [Gilks and Wild, 1992, Gorur and Teh, 2008], have proposed Adaptive Rejection Sampling (ARS) where, based on rejections, the q curve is updated to a lower curve q (cid:48) with a better acceptance rate.”