Reverend Bayes on Inference Engines: A Distributed Hierarchical Approach

By Judea Pearl. 1982

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

Determinantal Point Processes for Machine Learning (2012, background): “2 Belief Propagation We now describe the basic belief propagation algorithm, first introduced by Pearl [119].”
Learning Multiagent Communication with Backpropagation (2016, background): “The edg s w thin th graph rep ent the commun cation chann l b we n agents, with (3) being equivalent to belief pr pagation [22].”
Fast Gradient-Based Inference with Continuous Latent Variable Models in Auxiliary Form (2013, background): “thms for approximate inference in dynamic Bayesian networks can be roughly divided into two categories: sampling approaches and parametric approaches. Parametric approaches include Belief Propagation [Pea82] or the more recent Expectation Propagation (EP) [Min01]). When it is not reasonable or possi”
Learning to Reason: Leveraging Neural Networks for Approximate DNF Counting (2019, methodology): “Beyond hashing techniques, loopy belief propagation (LBP) (Pearl 1982; Murphy, Weiss, and Jordan 1999) has been applied to approximate WMC.”
Causal networks: semantics and expressiveness (2013, methodology): “Third, there exist various efficient algorithms that make implicit use of these representation schemes [6] [7] [8] [9] [12] [13].”
Bayesian Artificial Intelligence (2004, methodology): “Pearl [215] developed the message passing method for inference in simple trees.”
Factor Graphs for Robot Perception (2017, background): “In the AI community the development of expert systems spurred the development of probabilistic reasoning systems, and, in a landmark paper, Pearl [162] developed belief propagation in trees.”
Bayesian Networks and Decision Graphs (2001, background): “The connection between causation and conditional independence was studied by Spohn (1980), and later investigated with special focus on Bayesian networks in (Pearl, 2000).”
GraphChi: Large-Scale Graph Computation on Just a PC (2012, unknown): “Belief Propagation (BP) [39], is a powerful method based on iterative message passing between vertices.”
Efficient Learning for Crowdsourced Regression (2017, methodology): “We propose a novel inference algorithm based on belief propagation (BP) Pearl (1982) to estimate workers’ noise levels, and consequently estimate the ground truths.”
Opcode sequences as representation of executables for data-mining-based unknown malware detection (2013, background): “Bayesian networks [52] are probabilistic models for multivariate analysis.”
From certainty factors to belief networks (1992, background): “Pearl describes this phenomenon in detail [33, Chapter 1].”
Deep Structured Prediction with Nonlinear Output Transformations (2018, methodology): “There do exist choices for R and T , however, where global convergence guarantees can be attained – specifically, if R forms a tree [39] and if T is concave in y (which can be attained, for example, using an input-convex neural network [2]).”
Statistical physics of inference: thresholds and algorithms (2015, unknown): “There is a global connection between the Bethe approximation and the so-called belief propagation approach in computer science (in error correction [82] and Bayesian networks [189]) as was realized in connection to error correcting codes by Kabashima and Saad [119], and was put in a more general set”
Crowdsourced Data Management: A Survey (2016, methodology): “[64] model workers and tasks using a bipartite graph, and propose an iterative approach based on Belief Propagation [94] to derive workers’ models.”
Probabilistic reconstruction in compressed sensing: algorithms, phase diagrams, and threshold achieving matrices (2012, background): “It was discovered independently by several communities, in coding [26], in inference [27], or in statistical physics [12].”
Detection of Early-Stage Enterprise Infection by Mining Large-Scale Log Data (2014, background): “In typical implementations, the entire graph is constructed in advance ad “beliefs” are transmitted from every node to all its neighbors in each iteration.”
Supervised Community Detection with Line Graph Neural Networks (2017, background): “Belief propagation (BP) is a dynamical-programming-style algorithm for computing exactly or approximating marginal distributions in graphical model inferences (Pearl, 1982; Yedidia et al., 2003).”