On parallel implementation of sequential Monte Carlo methods: the island particle model
From MaRDI portal
Abstract: The approximation of the Feynman-Kac semigroups by systems of interacting particles is a very active research field, with applications in many different areas. In this paper, we study the parallelization of such approximations. The total population of particles is divided into sub-populations, referred to as emph{islands}. The particles within each island follow the usual selection / mutation dynamics. We show that the evolution of each island is also driven by a Feynman-Kac semigroup, whose transition and potential can be explicitly related to ones of the original problem. Therefore, the same genetic type approximation of the Feynman-Kac semi-group may be used at the island level; each island might undergo selection / mutation algorithm. We investigate the impact of the population size within each island and the number of islands, and study different type of interactions. We find conditions under which introducing interactions between islands is beneficial. The theoretical results are supported by some Monte Carlo experiments.
Recommendations
- Convergence properties of weighted particle islands with application to the double bootstrap algorithm
- Parallel and interacting Markov chain Monte Carlo algorithm
- Parallelizing particle filters with butterfly interactions
- Interacting sequential Monte Carlo samplers for trans-dimensional simulation
- On the role of interaction in sequential Monte Carlo algorithms
Cites work
- \(\mathrm{SMC}^2\): an efficient algorithm for sequential analysis of state space models
- A sequential particle filter method for static models
- Adaptive sequential posterior simulators for massively parallel computing environments
- Blind Deconvolution via Sequential Imputations
- scientific article; zbMATH DE number 2106098 (Why is no real title available?)
- Interacting sequential Monte Carlo samplers for trans-dimensional simulation
- Limit theorems for weighted samples with applications to sequential Monte Carlo methods
- Monte Carlo strategies in scientific computing
- On adaptive resampling strategies for sequential Monte Carlo methods
- Sequential Monte Carlo Methods in Practice
Cited in
(20)- A method for high-dimensional smoothing
- Parallelization of a dynamic Monte Carlo algorithm: A partially rejection-free conservative approach
- Compressed Monte Carlo with application in particle filtering
- Sequential ensemble transform for Bayesian inverse problems
- Parallel sequential Monte Carlo for stochastic gradient-free nonconvex optimization
- Sequential Bayesian inference for implicit hidden Markov models and current limitations
- Theory of segmented particle filters
- Learning undirected graphical models using persistent sequential Monte Carlo
- A pseudo-marginal sequential Monte Carlo algorithm for random effects models in Bayesian sequential design
- Forest resampling for distributed sequential Monte Carlo
- Parallel generalized elliptical slice sampling with adaptive regional pseudo-priors
- An ergodic theorem for the weighted ensemble method
- The Wigner branching random walk: efficient implementation and performance evaluation
- Point process-based Monte Carlo estimation
- A stable particle filter for a class of high-dimensional state-space models
- An Invitation to Sequential Monte Carlo Samplers
- On the role of interaction in sequential Monte Carlo algorithms
- Finite-sample complexity of sequential Monte Carlo estimators
- Generalized Bayesian multidimensional scaling and model comparison
- Inference on high-dimensional implicit dynamic models using a guided intermediate resampling filter
This page was built for publication: On parallel implementation of sequential Monte Carlo methods: the island particle model
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5962737)
