Stable local computation with conditional Gaussian distributions

By Steffen L. Lauritzen, Frank Jensen. 2001

297 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability, Signal Processing

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

1. Xintao Wu
2. Lu Zhang
3. Yongkai Wu
4. Vaishak Belle
5. Silvia Chiappa
6. J. Ramon
7. Jesse Davis
8. Anders L. Madsen

In Statistics and Probability

1. Yongkai Wu
2. Lu Zhang
3. Xintao Wu
4. J. Ramon
5. Anders L. Madsen
6. Vaishak Belle
7. Andrea Passerini
8. David Maxwell Chickering

In Signal Processing

1. J. Ramon
2. Jesse Davis
3. Anders L. Madsen
4. Vaishak Belle
5. Andrea Passerini
6. David Maxwell Chickering
7. Guy Van den Broeck
8. Tom Heskes Onno Zoeter

Citations in the corpus

• Causal Modeling-Based Discrimination Discovery and Removal: Criteria, Bounds, and Algorithms (2019, methodology): “A possible way is to employ the Conditional Linear Gaussian (CLG) distribution which is used in the Bayesian network to deal with the mixture of discrete and numerical variables [38].”
• Probabilistic Inference in Hybrid Domains by Weighted Model Integration (2015, background): “[2000]), or they make strong assumptions on the form of the potentials, such as Gaussian distributions [Lauritzen and Jensen, 2001].”
• Component Caching in Hybrid Domains with Piecewise Polynomial Densities (2016, methodology): “For hybrid graphical models, however, most inference algorithms are either approximate, e.g., (Murphy 1999; Gogate and Dechter 2005; Lunn et al. 2000), or they make strong assumptions on the form of the densities, e.g., Gaus-sians (Lauritzen and Jensen 2001).”
• Bayesian Networks and Influence Diagrams: A Guide to Construction and Analysis (2007, methodology): “Notice that since any ordinary influence diagram may be represented as a limited memory influence diagram, the SPU solution process may be used to solve influence diagrams; see, for example, Madsen & Nilsson (2001). Any ordinary influence diagram is a special case of a limited memory influence diagr”
• Learning relational dependency networks in hybrid domains (2015, background): “In terms of hybrid Bayesian networks, most of the work has focused on inference (Koller et al 1999; Yuan and Druzdzel 2007; Murphy 1998; Moral et al 2001; Lauritzen and Jensen 2001).”
• Expectation propagation for approximate inference in dynamic Bayesian networks (2002, background): “(Lauritzen and Jensen, 2001)), as well as the combination of expectation propagation with other (sam-”
• ARMA Time-Series Modeling with Graphical Models (2004, methodology): “By realizing that the σ ARMA xp model defines a Bayes net, the expected sufficient statistics involved in the above ML estimation can be efficiently computed by the procedure described in Lauritzen and Jensen (2001), or any other stable method for inference in Gaussian Bayes nets.”