A backward particle interpretation of Feynman-Kac formulae

Pierre Del Moral; Arnaud Doucet; Sumeetpal S. Singh

ESAIM: Mathematical Modelling and Numerical Analysis (2010)

  • Volume: 44, Issue: 5, page 947-975
  • ISSN: 0764-583X

Abstract

top
We design a particle interpretation of Feynman-Kac measures on path spaces based on a backward Markovian representation combined with a traditional mean field particle interpretation of the flow of their final time marginals. In contrast to traditional genealogical tree based models, these new particle algorithms can be used to compute normalized additive functionals “on-the-fly” as well as their limiting occupation measures with a given precision degree that does not depend on the final time horizon. We provide uniform convergence results w.r.t. the time horizon parameter as well as functional central limit theorems and exponential concentration estimates, yielding what seems to be the first results of this type for this class of models. We also illustrate these results in the context of filtering of hidden Markov models, as well as in computational physics and imaginary time Schroedinger type partial differential equations, with a special interest in the numerical approximation of the invariant measure associated to h-processes.

How to cite

top

Del Moral, Pierre, Doucet, Arnaud, and Singh, Sumeetpal S.. "A backward particle interpretation of Feynman-Kac formulae." ESAIM: Mathematical Modelling and Numerical Analysis 44.5 (2010): 947-975. <http://eudml.org/doc/250778>.

@article{DelMoral2010,
abstract = { We design a particle interpretation of Feynman-Kac measures on path spaces based on a backward Markovian representation combined with a traditional mean field particle interpretation of the flow of their final time marginals. In contrast to traditional genealogical tree based models, these new particle algorithms can be used to compute normalized additive functionals “on-the-fly” as well as their limiting occupation measures with a given precision degree that does not depend on the final time horizon. We provide uniform convergence results w.r.t. the time horizon parameter as well as functional central limit theorems and exponential concentration estimates, yielding what seems to be the first results of this type for this class of models. We also illustrate these results in the context of filtering of hidden Markov models, as well as in computational physics and imaginary time Schroedinger type partial differential equations, with a special interest in the numerical approximation of the invariant measure associated to h-processes. },
author = {Del Moral, Pierre, Doucet, Arnaud, Singh, Sumeetpal S.},
journal = {ESAIM: Mathematical Modelling and Numerical Analysis},
keywords = {Feynman-Kac models; mean field particle algorithms; functional central limit theorems; exponential concentration; non asymptotic estimates; functional central limit theorems; convergence filtering; hidden Markov models; Schrödinger equations; numerical example},
language = {eng},
month = {8},
number = {5},
pages = {947-975},
publisher = {EDP Sciences},
title = {A backward particle interpretation of Feynman-Kac formulae},
url = {http://eudml.org/doc/250778},
volume = {44},
year = {2010},
}

TY - JOUR
AU - Del Moral, Pierre
AU - Doucet, Arnaud
AU - Singh, Sumeetpal S.
TI - A backward particle interpretation of Feynman-Kac formulae
JO - ESAIM: Mathematical Modelling and Numerical Analysis
DA - 2010/8//
PB - EDP Sciences
VL - 44
IS - 5
SP - 947
EP - 975
AB - We design a particle interpretation of Feynman-Kac measures on path spaces based on a backward Markovian representation combined with a traditional mean field particle interpretation of the flow of their final time marginals. In contrast to traditional genealogical tree based models, these new particle algorithms can be used to compute normalized additive functionals “on-the-fly” as well as their limiting occupation measures with a given precision degree that does not depend on the final time horizon. We provide uniform convergence results w.r.t. the time horizon parameter as well as functional central limit theorems and exponential concentration estimates, yielding what seems to be the first results of this type for this class of models. We also illustrate these results in the context of filtering of hidden Markov models, as well as in computational physics and imaginary time Schroedinger type partial differential equations, with a special interest in the numerical approximation of the invariant measure associated to h-processes.
LA - eng
KW - Feynman-Kac models; mean field particle algorithms; functional central limit theorems; exponential concentration; non asymptotic estimates; functional central limit theorems; convergence filtering; hidden Markov models; Schrödinger equations; numerical example
UR - http://eudml.org/doc/250778
ER -

References

top
  1. D. Bakry, L'hypercontractivitée et son utilisation en théorie des semigroupes, in Lecture Notes in Math.1581, École d'été de St. Flour XXII, P. Bernard Ed. (1992).  
  2. P. Billingsley, Probability and Measure. Third edition, Wiley series in probability and mathematical statistics (1995).  Zbl0822.60002
  3. E. Cancès, B. Jourdain and T. Lelièvre, Quantum Monte Carlo simulations of fermions. A mathematical analysis of the fixed-node approximation. ESAIM: M2AN16 (2006) 1403–1449.  Zbl1098.81095
  4. F. Cerou, P. Del Moral and A. Guyader, A non asymptotic variance theorem for unnormalized Feynman-Kac particle models. Ann. Inst. Henri Poincaré (to appear).  Zbl1233.60047
  5. P.-A. Coquelin, R. Deguest and R. Munos, Numerical methods for sensitivity analysis of Feynman-Kac models. Available at , HAL-INRIA Research Report 6710 (2008).  URIhttp://hal.inria.fr/inria-00336203/en/
  6. D. Crisan, P. Del Moral and T. Lyons, Interacting Particle Systems. Approaximations of the Kushner-Stratonovitch Equation. Adv. Appl. Probab.31 (1999) 819–838.  Zbl0947.60040
  7. P. Del Moral, Feynman-Kac formulae. Genealogical and interacting particle systems with applications. Probability and its Applications, Springer Verlag, New York (2004).  
  8. P. Del Moral and A. Doucet, Particle motions in absorbing medium with hard and soft obstacles. Stoch. Anal. Appl.22 (2004) 1175–1207.  Zbl1071.60100
  9. P. Del Moral and L. Miclo, Branching and interacting particle systems approximations of Feynman-Kac formulae with applications to non-linear filtering, in Séminaire de Probabilités XXXIV, Lecture Notes in Math.1729, Springer, Berlin (2000) 1–145.  Zbl0963.60040
  10. P. Del Moral and L. Miclo, Particle approximations of Lyapunov exponents connected to Schroedinger operators and Feynman-Kac semigroups. ESAIM: PS7 (2003) 171–208.  Zbl1040.81009
  11. P. Del Moral and E. Rio, Concentration inequalities for mean field particle models. Available at , HAL-INRIA Research Report 6901 (2009).  URIhttp://hal.inria.fr/inria-00375134/fr/
  12. P. Del Moral, J. Jacod and P. Protter, The Monte Carlo Method for filtering with discrete-time observations. Probab. Theory Relat. Fields120 (2001) 346–368.  Zbl0979.62072
  13. P. Del Moral, A. Doucet and S.S. Singh, Forward smoothing using sequential Monte Carlo. Cambridge University Engineering Department, Technical Report CUED/F-INFENG/TR 638 (2009).  
  14. G.B. Di Masi, M. Pratelli and W.G. Runggaldier, An approximation for the nonlinear filtering problem with error bounds. Stochastics14 (1985) 247–271.  Zbl0566.60046
  15. R. Douc, A. Garivier, E. Moulines and J. Olsson, On the forward filtering backward smoothing particle approximations of the smoothing distribution in general state spaces models. Technical report, available at arXiv:0904.0316.  Zbl1237.60026
  16. A. Doucet, N. De Freitas and N. Gordon Eds., Sequential Monte Carlo Methods in Pratice. Statistics for engineering and Information Science, Springer, New York (2001).  
  17. M. El Makrini, B. Jourdain and T. Lelièvre, Diffusion Monte Carlo method: Numerical analysis in a simple case. ESAIM: M2AN41 (2007) 189–213.  Zbl1135.81379
  18. M. Émery, Stochastic calculus in manifolds. Universitext, Springer-Verlag, Berlin (1989).  Zbl0697.60060
  19. S.J. Godsill, A. Doucet and M. West, Monte Carlo smoothing for nonlinear time series. J. Am. Stat. Assoc.99 (2004) 156–168.  Zbl1089.62517
  20. N. Ikeda and S. Watanabe, Stochastic differential equations and diffusion processes24. Second edition, North-Holland Mathematical Library, North-Holland Publishing Co., Amsterdam (1989).  Zbl0684.60040
  21. M. Kac, On distributions of certain Wiener functionals. Trans. Am. Math. Soc.65 (1949) 1–13.  Zbl0032.03501
  22. G. Kallianpur and C. Striebel, Stochastic differential equations occurring in the estimation of continuous parameter stochastic processes. Tech. Rep. 103, Department of Statistics, University of Minnesota, Minneapolis (1967).  Zbl0195.44503
  23. N. Kantas, A. Doucet, S.S. Singh and J.M. Maciejowski, An overview of sequential Monte Carlo methods for parameter estimation in general state-space models, in Proceedings IFAC System Identification(SySid) Meeting, available at (2009).  URIhttp://publications.eng.cam.ac.uk/16156/
  24. H. Korezlioglu and W.J. Runggaldier, Filtering for nonlinear systems driven by nonwhite noises: an approximating scheme. Stoch. Stoch. Rep.44 (1983) 65–102.  Zbl0786.60058
  25. J. Picard, Approximation of the nonlinear filtering problems and order of convergence, in Filtering and control of random processes, Lecture Notes in Control and Inf. Sci.61, Springer (1984) 219–236.  
  26. G. Poyiadjis, A. Doucet and S.S. Singh, Sequential Monte Carlo computation of the score and observed information matrix in state-space models with application to parameter estimation. Technical Report CUED/F-INFENG/TR 628, Cambridge University Engineering Department (2009).  Zbl1214.62093
  27. D. Revuz, Markov chains. North-Holland (1975).  
  28. M. Rousset, On the control of an interacting particle approximation of Schroedinger ground states. SIAM J. Math. Anal.38 (2006) 824–844.  Zbl1174.60045
  29. A.N. Shiryaev, Probability, Graduate Texts in Mathematics95. Second edition, Springer (1986).  
  30. D.W. Stroock, Probability Theory: an Analytic View. Cambridge University Press, Cambridge (1994).  Zbl0960.60001
  31. D.W. Stroock, An Introduction to Markov Processes, Graduate Texts in Mathematics230. Springer (2005).  Zbl1068.60003

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.