Improved particle filter for nonlinear problems

By J. Carpenter, P. Clifford, Paul Fearnhead. 1999

Subfields: Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Probabilistic & Bayesian Methods

Citations in the corpus

Sequential Kernel Herding: Frank-Wolfe Optimization for Particle Filtering (2015, methodology): “Lower variance estimators can be obtained by using a diﬀerent resampling mechanism for the particles than this multinomial sampling scheme, such as stratiﬁed resampling (Carpenter et al., 1999) and are usually used in practice instead.”
Structured Pruning of Deep Convolutional Neural Networks (2015, methodology): “During the whole process, we set aside validation set for network convergence.”
Localization in highly dynamic environments using dual-timescale NDT-MCL (2014, methodology): “We use a basic resampling approach [19], which is triggered when the variance of the weights grows over a given threshold.”
Sequential Monte Carlo samplers (2002, methodology): “As noted in the transdimensional MCMC and SMC literatures (e.g. Green (2003), Carpenter et al. (1999), Doucet et al. (2000) and Pitt and Shephard (1999)) and in Section 3.3.2, a potentially good way to generate proposals in new dimensional spaces is to use the full conditional density.”
Resampling Methods for Particle Filtering: Classification, implementation, and strategies (2015, methodology): “It is known that multinomial resampling is not efficient [8], and this has motivated a search for faster methods.”
Active localization with dynamic obstacles (2016, background): “…done in state-of-the-art particle ﬁlters [20], the effective sample size ESS [21] is estimated after every update and, when it drops below 80% of the particle population, resampling [22] probabilistically eliminates particles with negligible weights and multiplies the ones with signiﬁcant weights.”
Negative association, ordering and convergence of resampling methods (2017, methodology): “The systematic resampling method of Carpenter et al. (1999), which involves a technique called universal sampling by Whitley (1994), is a very popular and computationally cheap resampling technique, with the property that the number of offspring of any sample with weight W in a population of N is…”
Particle filters for partially observed diffusions (2007, methodology): “Algorithms EPPF, RWPF1 and RWPF2 used the stratified resampling algorithm of Carpenter et al. (1999), with resampling at every iteration.”
Sequential Monte Carlo on large binary sampling spaces (2011, background): “There are, however, more efficient ways like residual (Liu and Chen 1998), stratified (Kitagawa 1996) and systematic resampling (Carpenter et al. 1999).”
Resampling algorithms and architectures for distributed particle filters (2005, methodology): “Resampling was proposed for use in particle filtering in various works including [2], [ 3 ], [13]–[16].”
Uncertainty in hydrologic modeling: Toward an integrated data assimilation framework (2007, methodology): “[77] For more details on the implementation and applications of particle filtering and various SMC methods, the readers are referred to Gordon et al. [1993], Carpenter et al. [1999] , Crisan et al. [1999], Doucet et al. [2001], Arulampalam et al. [2002], and Djuric ´e t al.[2003].”
SMC2: an efficient algorithm for sequential analysis of state space models (2011, background): “Other, more efficient schemes exist (Liu and Chen, 1998; Kitagawa, 1998; Carpenter et al., 1999), but are not discussed in the paper for the sake of simplicity.”
Quasi-Random Sampling for Condensation (2000, background): “Various techniques (such as importance sampling [2] and stratified sampling [ 9 ]) have been proposed to improve the efficiency of the representation.”
Adaptive methods for sequential importance sampling with application to state space models (2008, background): “This resampling procedure might however severely reduce the accuracy of the filter: Carpenter et al. (1999) give an example where the accuracy is reduced by a factor of 2; see also Douc et al. (2008) for a theoretical proof. consists in choosing a sensible risk criterion, which is not a…”