Semi-independent resampling for particle filtering
Computation
2018-02-14 v1
Abstract
Among Sequential Monte Carlo (SMC) methods,Sampling Importance Resampling (SIR) algorithms are based on Importance Sampling (IS) and on some resampling-based)rejuvenation algorithm which aims at fighting against weight degeneracy. However %whichever the resampling technique used this mechanism tends to be insufficient when applied to informative or high-dimensional models. In this paper we revisit the rejuvenation mechanism and propose a class of parameterized SIR-based solutions which enable to adjust the tradeoff between computational cost and statistical performances.
Cite
@article{arxiv.1710.05407,
title = {Semi-independent resampling for particle filtering},
author = {Roland Lamberti and Yohan Petetin and François Desbouvries and François Septier},
journal= {arXiv preprint arXiv:1710.05407},
year = {2018}
}