Group Importance Sampling for Particle Filtering and MCMC

Bayesian methods and their implementations by means of sophisticated Monte Carlo techniques have become very popular in signal processing over the last years. Importance Sampling (IS) is a well-known Monte Carlo technique that approximates integrals involving a posterior distribution by means of weighted samples. In this work, we study the assignation of a single weighted sample which compresses the information contained in a population of weighted samples. Part of the theory that we present as Group Importance Sampling (GIS) has been employed implicitly in different works in the literature. The provided analysis yields several theoretical and practical consequences. For instance, we discuss the application of GIS into the Sequential Importance Resampling framework and show that Independent Multiple Try Metropolis schemes can be interpreted as a standard Metropolis-Hastings algorithm, following the GIS approach. We also introduce two novel Markov Chain Monte Carlo (MCMC) techniques based on GIS. The first one, named Group Metropolis Sampling method, produces a Markov chain of sets of weighted samples. All these sets are then employed for obtaining a unique global estimator. The second one is the Distributed Particle Metropolis-Hastings technique, where different parallel particle filters are jointly used to drive an MCMC algorithm. Different resampled trajectories are compared and then tested with a proper acceptance probability. The novel schemes are tested in different numerical experiments such as learning the hyperparameters of Gaussian Processes, two localization problems in a wireless sensor network (with synthetic and real data) and the tracking of vegetation parameters given satellite observations, where they are compared with several benchmark Monte Carlo techniques. Three illustrative Matlab demos are also provided.

[1]  Jukka Corander,et al.  Layered adaptive importance sampling , 2015, Statistics and Computing.

[2]  Mathias Disney,et al.  Efficient Emulation of Radiative Transfer Codes Using Gaussian Processes and Application to Land Surface Parameter Inferences , 2016, Remote. Sens..

[3]  Luca Martino,et al.  Effective sample size for importance sampling based on discrepancy measures , 2016, Signal Process..

[4]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[5]  Roberto Casarin,et al.  Interacting multiple try algorithms with different proposal distributions , 2010, Statistics and Computing.

[6]  Fredrik Lindsten,et al.  High-Dimensional Filtering Using Nested Sequential Monte Carlo , 2016, IEEE Transactions on Signal Processing.

[7]  Anthony N. Pettitt,et al.  A Sequential Monte Carlo Algorithm to Incorporate Model Uncertainty in Bayesian Sequential Design , 2014 .

[8]  Alfred O. Hero,et al.  Relative location estimation in wireless sensor networks , 2003, IEEE Trans. Signal Process..

[9]  Petar M. Djuric,et al.  Resampling algorithms and architectures for distributed particle filters , 2005, IEEE Transactions on Signal Processing.

[10]  Daniel M. Roy,et al.  CONVERGENCE OF SEQUENTIAL MONTE CARLO-BASED SAMPLING METHODS , 2015 .

[11]  Eric Moulines,et al.  On parallel implementation of sequential Monte Carlo methods: the island particle model , 2013, Stat. Comput..

[12]  Luca Martino,et al.  A review of multiple try MCMC algorithms for signal processing , 2017, Digit. Signal Process..

[13]  Mónica F. Bugallo,et al.  Adaptive importance sampling in signal processing , 2015, Digit. Signal Process..

[14]  Nicholas G. Polson,et al.  Particle Filtering , 2006 .

[15]  Luca Martino,et al.  Cooperative parallel particle filters for online model selection and applications to urban mobility , 2015, Digit. Signal Process..

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

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

[18]  Xiaodong Wang,et al.  Monte Carlo methods for signal processing: a review in the statistical signal processing context , 2005, IEEE Signal Processing Magazine.

[19]  Jean-Marie Cornuet,et al.  Adaptive Multiple Importance Sampling , 2009, 0907.1254.

[20]  Mónica F. Bugallo,et al.  Heretical Multiple Importance Sampling , 2016, IEEE Signal Processing Letters.

[21]  A. Doucet,et al.  Particle Markov chain Monte Carlo methods , 2010 .

[22]  Luca Martino,et al.  A novel rejection sampling scheme for posterior probability distributions , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.

[23]  Xiaodong Wang,et al.  Monte Carlo methods for signal processing , 2005 .

[24]  D. Rubin.  Using the SIR algorithm to simulate posterior distributions , 1988 .

[25]  W H Wong,et al.  Dynamic weighting in Monte Carlo and optimization. , 1997, Proceedings of the National Academy of Sciences of the United States of America.

[26]  Jukka Corander,et al.  Orthogonal parallel MCMC methods for sampling and optimization , 2015, Digit. Signal Process..

[27]  Eric Moulines,et al.  Scaling analysis of multiple-try MCMC methods , 2012 .

[28]  David Luengo,et al.  Generalized Multiple Importance Sampling , 2015, Statistical Science.

[29]  Luca Martino,et al.  Group metropolis sampling , 2017, 2017 25th European Signal Processing Conference (EUSIPCO).

[30]  John W. Fisher,et al.  Nonparametric belief propagation for self-localization of sensor networks , 2005, IEEE Journal on Selected Areas in Communications.

[31]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[32]  Theodore S. Rappaport,et al.  Wireless communications - principles and practice , 1996 .

[33]  R. Carroll,et al.  Advanced Markov Chain Monte Carlo Methods: Learning from Past Samples , 2010 .

[34]  Rong Chen,et al.  Monte Carlo Bayesian Signal Processing for Wireless Communications , 2002, J. VLSI Signal Process..

[35]  Ben Calderhead,et al.  A general construction for parallelizing Metropolis−Hastings algorithms , 2014, Proceedings of the National Academy of Sciences.

[36]  A. Doucet,et al.  A Tutorial on Particle Filtering and Smoothing: Fifteen years later , 2008 .

[37]  William J. Fitzgerald,et al.  Markov chain Monte Carlo methods with applications to signal processing , 2001, Signal Process..

[38]  Fredrik Lindsten,et al.  Nested Sequential Monte Carlo Methods , 2015, ICML.

[39]  Joaquín Míguez,et al.  A proof of uniform convergence over time for a distributed particle filter , 2015, Signal Process..

[40]  Andreas M. Ali,et al.  An Empirical Study of Collaborative Acoustic Source Localization , 2007, 2007 6th International Symposium on Information Processing in Sensor Networks.

[41]  Mónica F. Bugallo,et al.  Sequential Monte Carlo methods under model uncertainty , 2016, 2016 IEEE Statistical Signal Processing Workshop (SSP).

[42]  Luca Martino,et al.  Fully adaptive Gaussian mixture Metropolis-Hastings algorithm , 2012, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[43]  Timothy J. Robinson,et al.  Sequential Monte Carlo Methods in Practice , 2003 .

[44]  François Septier,et al.  An improved SIR-based Sequential Monte Carlo algorithm , 2016, 2016 IEEE Statistical Signal Processing Workshop (SSP).

[45]  James V. Candy,et al.  Bayesian Signal Processing: Classical, Modern and Particle Filtering Methods , 2009 .

[46]  Jukka Corander,et al.  MCMC-Driven Adaptive Multiple Importance Sampling , 2015 .

[47]  Jesse Read,et al.  A distributed particle filter for nonlinear tracking in wireless sensor networks , 2014, Signal Process..

[48]  Petar M. Djuric,et al.  Adaptive Importance Sampling: The past, the present, and the future , 2017, IEEE Signal Processing Magazine.

[49]  Dani Gamerman,et al.  Markov Chain Monte Carlo: Stochastic Simulation for Bayesian Inference , 1997 .

[50]  Tim Hesterberg,et al.  Monte Carlo Strategies in Scientific Computing , 2002, Technometrics.

[51]  P. Djurić,et al.  Particle filtering , 2003, IEEE Signal Process. Mag..

[52]  J. Read,et al.  A multi-point Metropolis scheme with generic weight functions , 2011, 1112.4048.

[53]  L. Martino,et al.  Issues in the Multiple Try Metropolis mixing , 2017, Comput. Stat..

[54]  Alfred O. Hero,et al.  A Survey of Stochastic Simulation and Optimization Methods in Signal Processing , 2015, IEEE Journal of Selected Topics in Signal Processing.

[55]  J. Propp,et al.  Exact sampling with coupled Markov chains and applications to statistical mechanics , 1996 .

[56]  K OrJ.  Numerical Bayesian methods applied to signal processing , 1996 .

[57]  Luca Martino,et al.  On the flexibility of the design of multiple try Metropolis schemes , 2012, Computational Statistics.

[58]  H. Haario,et al.  An adaptive Metropolis algorithm , 2001 .

[59]  P. Moral,et al.  Convergence properties of weighted particle islands with application to the double bootstrap algorithm , 2014, 1410.4231.

[60]  J. Chen,et al.  Defining leaf area index for non‐flat leaves , 1992 .

[61]  Mónica F. Bugallo,et al.  Efficient Multiple Importance Sampling Estimators , 2015, IEEE Signal Processing Letters.

[62]  Christiane Lemieux,et al.  Acceleration of the Multiple-Try Metropolis algorithm using antithetic and stratified sampling , 2007, Stat. Comput..

[63]  Mark C. Reed,et al.  Advanced Markov Chain Monte Carlo Methods for Iterative (Turbo) Multiuser Detection , 2006 .

[64]  D. Rubin,et al.  Inference from Iterative Simulation Using Multiple Sequences , 1992 .

[65]  Luca Martino,et al.  Weighting a resampled particle in Sequential Monte Carlo , 2016, 2016 IEEE Statistical Signal Processing Workshop (SSP).

[66]  Fabrizio Leisen,et al.  ON MULTIPLE TRY SCHEMES AND THE PARTICLE METROPOLIS-HASTINGS ALGORITHM , 2014 .

[67]  Moon Gi Kang,et al.  Super-resolution image reconstruction , 2010, 2010 International Conference on Computer Application and System Modeling (ICCASM 2010).

[68]  Nando de Freitas,et al.  An Introduction to MCMC for Machine Learning , 2004, Machine Learning.

[69]  Anthony Lee,et al.  On the role of interaction in sequential Monte Carlo algorithms , 2013, 1309.2918.

[70]  Hoon Kim,et al.  Monte Carlo Statistical Methods , 2000, Technometrics.