On sequential Monte Carlo sampling methods for Bayesian filtering

By Arnaud Doucet, Simon J. Godsill, Christophe Andrieu. 2000

4,949 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability, 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 Artificial Intelligence

In Statistics and Probability

In Probabilistic & Bayesian Methods

Citations in the corpus

Sequential Kernel Herding: Frank-Wolfe Optimization for Particle Filtering (2015, methodology): “This poses diﬃculties for particle-ﬁltering-based solutions for two reasons: (1) the computational bottleneck arising from the likelihood evaluation implies that we cannot simply increase the number of particles to improve the accuracy, and (2) this type of “complicated” observation models will…”
Bias Correction of Learned Generative Models using Likelihood-Free Importance Weighting (2019, methodology): “While exact sampling from p✓, is intractable, we can instead perform sample from a particle-based approximation to p✓, via sampling-importance-resampling [25, 26] (SIR).”
Edward: A library for probabilistic modeling, inference, and criticism (2016, methodology): “The approach of Bayesian filtering (Doucet et al., 2000; Broderick et al., 2013) can be applied in Edward using recursive posterior inferences; the approach of population posteriors (McInerney et al.”
Gaussian particle filtering (2001, methodology): “Recently, importance sampling-based filters have been used to update the posterior distributions [20]–[23].”
Monte Carlo filtering for multi target tracking and data association (2005, background): “A full discussion of degeneracy and resampling falls outside the scope of this work, but more detail can be found in [10].”
Real-time visual tracking using compressive sensing (2011, methodology): “The proffered algorithms are conducted on 10 video sequences and we compare the `1 tracker, the Kernel Mean Shift (KMS) tracker [18] and a color-based PF tracker [11].”
Generalized Multiple Importance Sampling (2015, methodology): “In particular, as a rule of thumb, the following expression is commonly is used ÊSS = 1 ∑N n=1 w̄ 2 n , (40) where w̄n = wn ∑N k=1 wk is the n-th normalized weight [18, 19].”
An overview of sequential Monte Carlo methods for parameter estimation in general state-space models (2009, background): “The “optimal” choice is given by pθ (x1| y1) and pθ (xn| yn, xn−1) and in practice it is recommended to approximate these distributions if it is not possible to sample from them; see [16, 41] for various approximation strategies.”
Uncertainty in hydrologic modeling: Toward an integrated data assimilation framework (2007, methodology): “…al. [2002] for detailed derivation): w i *ð Þ kþ1 ¼ w i *ð Þ k p zkþ1jxikþ1 p xikþ1jxik q xikþ1jxik ; zkþ1 i ¼ 1; ; n: ð39Þ [72] Choice of an appropriate proposal importance density is crucial in the SIS algorithm as reported by several studies [e.g., Doucet et al., 2000; Arulampalam et al., 2002].”
An ensemble model for predicting the remaining useful performance of lithium-ion batteries (2013, background): “The sample xk is drawn from importance density q(xk|y1:k) based on the importance sampling [27,28].”
Uncertainty assessment of hydrologic model states and parameters: Sequential data assimilation using the particle filter (2005, methodology): “3.4. Problem: One-Dimensional State-Space Model [21] Consider the application of SIS and SIR particle filtering to the following non-Gaussian, nonlinear statespace and observation model used as an illustrative example by several authors [Andrade Netto et al., 1978; Gordon et al., 1993; Kitagawa, 199”
Adaptive Multiple Importance Sampling (2009, methodology): “Those methods are called adaptive importance sampling (AIS), but they also relate to particle filters (Gordon et al., 1993; Doucet et al., 2001) and sequential Monte Carlo methods (Doucet et al., 2000; Chopin, 2002; Del Moral et al., 2006).”
Marginalized particle filters for mixed linear/nonlinear state-space models (2005, methodology): “A recent important contribution is to use simulation based methods from mathematical statistics, sequential Monte Carlo methods, commonly referred to as particle filters [11], [12], [16].”
An Introduction to MCMC for Machine Learning (2004, background): “Other application areas include dynamic Bayesian networks (Doucet et al., 2000), conditionally Gaussian models (Carter & Kohn, 1994; De Jong & Shephard, 1995; Doucet, 1998) and model averaging for graphical models (Friedman & Koller, this issue).”
A Tutorial on Particle Filtering and Smoothing: Fifteen years later (2008, methodology): “In particular the use of standard suboptimal filtering techniques such as the Extended Kalman Filter or the Unscented Kalman Filter to obtain importance distributions is very popular in the literature [14, 37].”
Cooperative Localization in Wireless Networks (2009, methodology): “Note that messages may be represented in other ways, using techniques such as th e extended Kalman filter [125], [126] or Rao-Blackwellizatio n [98], [127].”