Auxiliary Variable Methods for Markov Chain Monte Carlo with Applications

Abstract Suppose that one wishes to sample from the density π(x) using Markov chain Monte Carlo (MCMC). An auxiliary variable u and its conditional distribution π(u|x) can be defined, giving the joint distribution π(x, u) = π(x)π(u|x). A MCMC scheme that samples over this joint distribution can lead to substantial gains in efficiency compared to standard approaches. The revolutionary algorithm of Swendsen and Wang is one such example. Besides reviewing the Swendsen-Wang algorithm and its generalizations, this article introduces a new auxiliary variable method called partial decoupling. Two applications in Bayesian image analysis are considered: a binary classification problem in which partial decoupling out performs Swendsen-Wang and single-site Metropolis methods, and a positron emission tomography (PET) reconstruction that uses the gray level prior of Geman and McClure. A generalized Swendsen–Wang algorithm is developed for this problem, which reduces the computing time to the point where MCMC is a viab...

[1]  E. Ising.  Beitrag zur Theorie des Ferromagnetismus , 1925 .

[2]  R. B. Potts.  Some generalized order-disorder transformations , 1952, Mathematical Proceedings of the Cambridge Philosophical Society.

[3]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[4]  J. M. Hammersley,et al.  Conditional Monte Carlo , 1956, JACM.

[5]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[6]  L. Shepp,et al.  Maximum Likelihood Reconstruction for Emission Tomography , 1983, IEEE Transactions on Medical Imaging.

[7]  Y. Ogata,et al.  Likelihood Analysis of Spatial Point Patterns , 1984 .

[8]  L. Shepp,et al.  A Statistical Model for Positron Emission Tomography , 1985 .

[9]  Wang,et al.  Nonuniversal critical dynamics in Monte Carlo simulations. , 1987, Physical review letters.

[10]  Stuart Geman,et al.  Statistical methods for tomographic image reconstruction , 1987 .

[11]  Ron,et al.  Simulations without critical slowing down. , 1988, Physical review letters.

[12]  A. Sokal,et al.  Generalization of the Fortuin-Kasteleyn-Swendsen-Wang representation and Monte Carlo algorithm. , 1988, Physical review. D, Particles and fields.

[13]  A. Raftery,et al.  Ice Floe Identification in Satellite Images Using Mathematical Morphology and Clustering about Principal Curves , 1992 .

[14]  T R Miller,et al.  Fast maximum-likelihood reconstruction. , 1992, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[15]  J. Besag,et al.  Spatial Statistics and Bayesian Computation , 1993 .

[16]  Peter Green,et al.  Spatial statistics and Bayesian computation (with discussion) , 1993 .

[17]  C. Geyer.  Estimating Normalizing Constants and Reweighting Mixtures , 1994 .

[18]  L. Tierney.  Markov Chains for Exploring Posterior Distributions , 1994 .

[19]  Kerrie Mengersen,et al.  [Bayesian Computation and Stochastic Systems]: Rejoinder , 1995 .

[20]  J. Besag,et al.  Bayesian Computation and Stochastic Systems , 1995 .

[21]  P. Green,et al.  Bayesian Computation and Stochastic , 1995 .

[22]  Merrilee Hurn.  Difficulties in the use of auxiliary variables in Markov chain Monte Carlo methods , 1997, Stat. Comput..

[23]  Paul. Damien.  Gibbs sampling for Bayesian nonconjugate models using auxilliary variables , 1997 .

[24]  A. Sokal.  Monte Carlo Methods in Statistical Mechanics: Foundations and New Algorithms , 1997 .

[25]  Antonietta Mira,et al.  On the use of auxiliary variables in Markov chain Monte Carlo sampling , 1997 .

[26]  Xiao-Li Meng,et al.  Simulating Normalizing Constants: From Importance Sampling to Bridge Sampling to Path Sampling , 1998 .

[27]  G. S. Fishman.  An Analysis of Swendsen–Wang and Related Sampling Methods , 1999 .

[28]  J. Rosenthal,et al.  Convergence of Slice Sampler Markov Chains , 1999 .