Dynamic bayesian networks: representation, inference and learning

By Kevin P. Murphy, Stuart J. Russell. 2002

3,006 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/5e86e17d83c97dafa3413d1d0dae219bd527ed61

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

Overview on Bayesian networks applications for dependability, risk analysis and maintenance areas (2012, background): “Their efficiency depends on the model complexity (Murphy, 2002).”
a CAPpella: programming by demonstration of context-aware applications (2004, methodology): “For its machine learning system, a CAPpella uses a Dynamic Bayesian Network (DBN) framework [18], a”
Modeling Purposeful Adaptive Behavior with the Principle of Maximum Causal Entropy (2010, background): “A dynamic Bayesian network (Murphy, 2002), DBN = (G,P ) is defined by: • A template directed graphical structure, G; and • A set of conditional probability distributions defining the relationship for P (Xt+1|Xt) (where Xt = {Xt,1, .”
Sigma-point kalman filters for probabilistic inference in dynamic state-space models (2004, background): “Dynamic state-space models (as defined in this thesis) can be interpreted as a special case of the more general framework of Bayesian networks [140] (also known as Bayes nets), which in turn is the combination of Bayesian probability theory [89] and graphical models [93].”
Unsupervised Learning (2018, unknown): “A generalization of these ideas is the notion of a dynamical Bayesian network (DBN) [32].”
Dynamic Graphical Models (2010, methodology): “Since the separator was optimized, the notion of the left or right interface [12], [4], [28], [19], [3] is no longer relevant since after the aforementioned max-flow optimization they are identical.”
A CCELERATING O NLINE R EINFORCEMENT L EARNING VIA M ODEL -B ASED M ETA -L EARNING (2021, background): “Methods range from probabilistic ﬁltering (Murphy & Russell, 2002; Hoffman et al., 2010; Broderick et al., 2013) to online gradient updates (Duchi et al., 2011).”
Bayesian Time Series Models: Expectation maximisation methods for solving (PO)MDPs and optimal control problems (2011, methodology): “Examples are message passing algorithms (loopy Belief Propagation, Expectation Propagation), variational approaches, approximate belief representations (particles, Assumed Density Filtering, Boyen-Koller) and arithmetic compilation (see, e.g., Minka, 2001; Murphy, 2002; Chavira et al., 2006).”
A Data-Driven Failure Prognostics Method Based on Mixture of Gaussians Hidden Markov Models (2012, methodology): “The inference complexity is given by the formulae (Forward-Backward and Viterbi) [32], [35].”
Fusion Techniques for Reliable Information: A Survey (2010, background): “The dynamic BN, which considers temporal dimensions, is more useful theory in pervasive computing areas [69].”
Informed Machine Learning – A Taxonomy and Survey of Integrating Prior Knowledge into Learning Systems (2019, background): “In other words, data assimilation problems are modelled using dynamic Bayesian networks [94] with continuous physically interpretable state spaces where transition kernels and observation operators reflect knowledge about the system dynamics, e.”
A Review on Video-Based Human Activity Recognition (2013, methodology): “A DBN [78] is a Bayesian network with the same structure unrolled in the time axis as shown in Figure 15 [7].”
Finding your friends and following them to where you are (2012, background): “Dynamic Bayesian networks (DBNs) are generative probabilistic graphical models of sequential data [21].”
Generating Text with Recurrent Neural Networks (2011, background): “This is a much easier learning task and it works surprisingly well provided the recurrent connections (1)In contrast, Dynamic Bayes Networks (Murphy, 2002), the probabilistic analogues of RNNs, do not have an efficient algorithm for computing their gradients.”