Slice Sampling

In many applications of artificial intelligence it is of interest to be able to generate samples from probability distributions. Since it is in most cases not possible to sample from a target distribution directly, especially if it has a very complex structure, special techniques are required that allow sampling from such distributions. One of these sampling techniques are called are Markov chain Monte Carlo(MCMC) methods that form one of the most important tools for such sampling purposes. The goal of this paper is to present one of these MCMC methods called slice sampling [1] and evaluate it in comparison to a modified version of it called elliptical Slice sampling [4] and the more popular Metropolis-Hastings method.

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

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

[3]  S. Adler.  Over-relaxation method for the Monte Carlo evaluation of the partition function for multiquadratic actions , 1981 .

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

[5]  Adrian F. M. Smith,et al.  Sampling-Based Approaches to Calculating Marginal Densities , 1990 .

[6]  P. Barone,et al.  Improving Stochastic Relaxation for Gussian Random Fields , 1990, Probability in the Engineering and Informational Sciences.

[7]  A. Horowitz.  A generalized guided Monte Carlo algorithm , 1991 .

[8]  P. Green,et al.  Metropolis Methods, Gaussian Proposals and Antithetic Variables , 1992 .

[9]  Radford M. Neal.  An improved acceptance procedure for the hybrid Monte Carlo algorithm , 1992, hep-lat/9208011.

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

[11]  W. Gilks,et al.  Adaptive Rejection Metropolis Sampling Within Gibbs Sampling , 1995 .

[12]  Geoffrey E. Hinton,et al.  Bayesian Learning for Neural Networks , 1995 .

[13]  Brendan J. Frey,et al.  Continuous Sigmoidal Belief Networks Trained using Slice Sampling , 1996, NIPS.

[14]  Walter R. Gilks,et al.  Corrigendum: Adaptive Rejection Metropolis Sampling , 1997 .

[15]  Radford M. Neal,et al.  Suppressing Random Walks in Markov Chain Monte Carlo Using Ordered Overrelaxation , 1995, Learning in Graphical Models.

[16]  D. Higdon.  Auxiliary Variable Methods for Markov Chain Monte Carlo with Applications , 1998 .

[17]  Antonietta Mira,et al.  Ordering, Slicing And Splitting Monte Carlo Markov Chains , 1998 .

[18]  Andrew Gelman,et al.  General methods for monitoring convergence of iterative simulations , 1998 .

[19]  L Tierney,et al.  Some adaptive monte carlo methods for Bayesian inference. , 1999, Statistics in medicine.

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

[21]  Andrew Thomas,et al.  WinBUGS - A Bayesian modelling framework: Concepts, structure, and extensibility , 2000, Stat. Comput..

[22]  P. Green,et al.  Delayed rejection in reversible jump Metropolis–Hastings , 2001 .

[23]  Radford M. Neal.  Slice Sampling , 2003, The Annals of Statistics.