Particle Belief Propagation

By A. Ihler, David A. McAllester. 2009

155 citations recorded by the source corpus.

How researchers see this paper

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

Citations in the corpus

Kernel Mean Embedding of Distributions: A Review and Beyonds (2016, methodology): “Ihler and McAllester (2009) considers a particle belief propagation (BP) where the messages are expressed as a function of a distribution of particles at each node. Unlike these algorithms, the embedding-based algorithm proposed in Song et al. (2010b) expresses the message mts(s) between pairs of no”
Kernel Belief Propagation (2011, methodology): “We describe two competing approaches for nonparametric belief propagation: Gaussian mixture BP, originally known as non-parametric BP (Sudderth et al., 2003), and particle BP (Ihler & McAllester, 2009).”
Kernel Embeddings of Conditional Distributions: A Unified Kernel Framework for Nonparametric Inference in Graphical Models (2013, methodology): “3 Experimental Results We compare with two state-of-the-art approaches to nonparametric belief propagation: Gaussian Mixture BP [2] and Particle BP [26].”
Provable Bayesian Inference via Particle Mirror Descent (2015, methodology): “[37] and [21, 25] extend the belief propagation algorithm with nonparametric models by kernel embedding and particle approximation, respectively.”
Harmony Potentials (2011, background): “2000; Coughlan and Ferreira 2002) or sampling the label space (Ihler and McAllester 2009; Koller et al. 1999; Sudderth et al. 2002).”
Kernel Embeddings of Latent Tree Graphical Models (2011, background): “Furthermore, we choose the bandwidth σ for the Gaussian RBF kernel needed for the covariance operators using median distance between pairs of training points.”
Semantic Multi-view Stereo: Jointly Estimating Objects and Voxels (2017, methodology): “For the pose variables p, we follow a particle based strategy [18] and maintain a sample distribution {p((1)) s , .”
Continuous Markov Random Fields for Robust Stereo Estimation (2012, result): “In contrast, some previous works used particle belief propagation (PBP) [33, 34, 10] which correspond to non-convex norm-product with the Bethe entropy approximation.”
Inference in Probabilistic Graphical Models by Graph Neural Networks (2018, background): “Furthermore, the basic update steps in belief propagation may not have efﬁcient or even closed-form solutions, leading researchers to construct BP variants [3, 4, 5] or generalizations [6].”
Patches, Planes and Probabilities: A Non-Local Prior for Volumetric 3D Reconstruction (2016, methodology): “To tackle these challenges, we exploit particle belief propagation [24], which has been adopted by many works with great success [4, 19, 36, 53].”
Discrete-Continuous Depth Estimation from a Single Image (2014, methodology): “We make use of particle belief propagation [12, 20] to perform inference in the resulting higher-order, discrete-continuous graphical model.”
Nonparametric Tree Graphical Models via Kernel Embeddings (2010, methodology): “Unlike Ihler & McAllester (2009), for example, who assume that the node and edge potentials are given in advance and that they can sample from certain distributions associated with these potentials , we learn the model representation from data, do not require sampling in the inference procedure,…”
Bayesian Probabilistic Numerical Methods (2017, methodology): “…more general approach, sequential Monte Carlo methods can be used to propagate a collection of particles through the pipeline P , similar to work on SMC for general graphical models (Briers et al., 2005; Ihler and McAllester, 2009; Lienart et al., 2015; Lindsten et al., 2017; Paige and Wood, 2015).”
Proximal Deep Structured Models (2016, methodology): “Exact inference is possible for a certain type of MRFs, i.e., Gaussian MRFs with positive semidefinite precision matrix.”