Advanced Markov Chain Monte Carlo Methods: Learning from Past Samples

244 citations recorded by the source corpus.

How researchers see this paper

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

A Theory of Generative ConvNet (2016, methodology): “(3) For generative models, we usually need to use Markov chain Monte Carlo (MCMC) [20, 19] to generate images from the models, but the convergence properties of the MCMC algorithms are unknown.”
A Structural Model of Segregation in Social Networks (2010, result): “Fourth, we combine previous results to prove that the approximate algorithm converges to the correct posterior A similar proof strategy is contained in Liang et al. (2010) and Andrieu and Roberts (2009).”
On the flexibility of the design of multiple try Metropolis schemes (2012, methodology): “A very powerful class of Monte Carlo techniques is the so-called Markov Chain Monte Carlo (MCMC) algorithms (Gamerman and Lopes 2006; Gilks et al. 1995; Liang et al. 2010; Liu 2004; Robert and Casella 2004).”
MCMC-Driven Adaptive Multiple Importance Sampling (2015, background): “with means ν1 = [−10,−10]&, ν2 = [0, 16]&, ν3 = [13, 8]&, ν4 = [−9, 7]&, ν5 = [14,−14]&, and covariance matrices Σ1 = [2, 0.”
Bayesian Reinforcement Learning in Factored POMDPs (2018, methodology): “This particular combination of MCMC methods — Metropolis-Hastings in one of Gibbs’s conditional sampling steps — is also referred to as MH-within-Gibbs and, has been known to converge to the true distribution even if the Metropolis-Hastings part only consist of one sample per step [23, 24, 27, 36, 4”