On variance stabilisation by double Rao-Blackwellisation
نویسندگان
چکیده
Population Monte Carlo has been introduced as a sequential importance sampling technique to overcome poor fit of the importance function. In this paper, we compare the performances of the original Population Monte Carlo algorithm with a modified version that eliminates the influence of the transition particle via a double Rao-Blackwellisation. This modification is shown to improve the exploration of the modes through an large simulation experiment on posterior distributions of mean mixtures of distributions. Key-words: Importance Sampling mixture, adaptive Monte Carlo, Population Monte Carlo, multimodality, mixture of distributions, random walk. ∗ CEREMADE, Université Paris Dauphine † INRIA Saclay, Projet select, Université Paris-Sud & CREST, INSEE ‡ CEREMADE, Université Paris Dauphine & CREST, INSEE in ria -0 02 60 14 1, v er si on 1 3 M ar 2 00 8 Schémas Population Monte Carlo et double Rao-Blackwellisation Résumé : Les algorithmes Population Monte Carlo (PMC) sont des schémas d’échantillonnage préférentiel adaptatifs qui ont été introduits afin d’ajuster la distribution d’importance à la loi cible. Dans cet article, nous comparons les performances du schéma PMC classique à celles d’une version modifiée qui diminue la variabilité des poids d’importance en utilisant une technique de double Rao-Blackwellisation. Il est montré que cette modification permet d’améliorer significativement les capacités d’exploration de l’algorithme dans le cas de lois cibles multi-modales. Mots-clés : Échantillonnage préférentiel, mélanges de lois, méthodes de Monte-Carlo adaptatives, marches aléatoires. in ria -0 02 60 14 1, v er si on 1 3 M ar 2 00 8 PMC schemes and double Rao-Blackwellisation 3
منابع مشابه
An explicit variance reduction expression for the Rao-Blackwellised particle filter
Particle filters (PFs) have shown to be very potent tools for state estimation in nonlinear and/or non-Gaussian state-space models. For certain models, containing a conditionally tractable substructure (typically conditionally linear Gaussian or with finite support), it is possible to exploit this structure in order to obtain more accurate estimates. This has become known as Rao-Blackwellised p...
متن کاملAn explicit variance reduction expression for the Rao-Blackwellised particle filter, Report no. LiTH-ISY-R-2979
Particle lters (PFs) have shown to be very potent tools for state estimation in nonlinear and/or non-Gaussian state-space models. For certain models, containing a conditionally tractable substructure (typically conditionally linear Gaussian or with nite support), it is possible to exploit this structure in order to obtain more accurate estimates. This has become known as Rao-Blackwellised parti...
متن کاملRao-Blackwellised Particle Filter with Adaptive System Noise and its Evaluation for Tracking in Surveillance
In the visual tracking domain, Particle Filtering (PF) can become quite inefficient when being applied into high dimensional state space. Rao-Blackwellisation [1] has been shown to be an effective method to reduce the size of the state space by marginalizing out some of the variables analytically . In this paper based on our previous work [3] we propose an RBPF tracking algorithm with adaptive ...
متن کاملRao-Blackwellised Particle Filtering via Data Augmentation
Arnaud Doucet EE Engineering University of Melbourne Parkville, Victoria 3052 Australia [email protected] In this paper, we extend the Rao-Blackwellised particle filtering method to more complex hybrid models consisting of Gaussian latent variables and discrete observations. This is accomplished by augmenting the models with artificial variables that enable us to apply Rao-Blackwellisation. Ot...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2008