Markov Chain Truncation for Doubly-Intractable Inference

Computing partition functions, the normalizing constants of probability distributions, is often hard. Variants of importance sampling give unbiased estimates of a normalizer Z, however, unbiased estimates of the reciprocal 1/Z are harder to obtain. Unbiased estimates of 1/Z allow Markov chain Monte Carlo sampling of "doubly-intractable" distributions, such as the parameter posterior for Markov Random Fields or Exponential Random Graphs. We demonstrate how to construct unbiased estimates for 1/Z given access to black-box importance sampling estimators for Z. We adapt recent work on random series truncation and Markov chain coupling, producing estimators with lower variance and a higher percentage of positive estimates than before. Our debiasing algorithms are simple to implement, and have some theoretical and empirical advantages over existing methods.

[1]  Zoubin Ghahramani,et al.  MCMC for Doubly-intractable Distributions , 2006, UAI.

[2]  A. Doucet,et al.  Efficient implementation of Markov chain Monte Carlo when using an unbiased likelihood estimator , 2012, 1210.1871.

[3]  G. Casella,et al.  Rao-Blackwellisation of sampling schemes , 1996 .

[4]  P. Jacob,et al.  On nonnegative unbiased estimators , 2013, 1309.6473.

[5]  Peter W. Glynn,et al.  Exact estimation for Markov chain equilibrium expectations , 2014, Journal of Applied Probability.

[6]  Radford M. Neal Annealed importance sampling , 1998, Stat. Comput..

[7]  C. Andrieu,et al.  The pseudo-marginal approach for efficient Monte Carlo computations , 2009, 0903.5480.

[8]  David Wilson,et al.  Coupling from the past: A user's guide , 1997, Microsurveys in Discrete Probability.

[9]  Yves F. Atchad'e,et al.  On Russian Roulette Estimates for Bayesian Inference with Doubly-Intractable Likelihoods , 2013, 1306.4032.

[10]  Raymond L. Murray,et al.  Particle-transport simulation with the Monte Carlo method By and . Technical Information Center, ERDA, 1975, (TID-26607) 115 pp. $dollar;5.45 , 1977 .

[11]  Don McLeish,et al.  A general method for debiasing a Monte Carlo estimator , 2010, Monte Carlo Methods Appl..

[12]  Alberto Caimo,et al.  Bayesian inference for exponential random graph models , 2010, Soc. Networks.

[13]  J. Møller,et al.  An efficient Markov chain Monte Carlo method for distributions with intractable normalising constants , 2006 .

[14]  I. Lux Monte Carlo Particle Transport Methods: Neutron and Photon Calculations , 1991 .

[15]  Iain Murray,et al.  Pseudo-Marginal Slice Sampling , 2015, AISTATS.