Inference in belief networks: A procedural guide

Huang, Cecil; Darwiche, Adnan

doi:10.1016/s0888-613x(96)00069-2

Cited by 313 publications

(221 citation statements)

References 13 publications

(16 reference statements)

Supporting

Mentioning

214

Contrasting

Unclassified

Order By: Relevance

“…If the graph G is not chordal, one adds in a set of additional edges, which are called fill-in edges, to make the graph chordal. Even though the problem of finding the fill-in with the minimum number of edges is NP-complete (see Yannakakis (1981)), there are efficient algorithms to find fill-ins with reasonably small number of edges (see for example, Huang and Darwiche (1996) and Natanzon et al (2000)). …”

Section: Construction Of Regular Coversmentioning

confidence: 99%

“…In our experiments, we use the minweightElimOrder function in PMTK3, a Matlab toolkit for probabilistic modeling (see Dunham and Murphy (2012)), which is based on a fill-in algorithm developed by Huang and Darwiche (1996). One can then construct a regular cover E from the modified chordal graph using the L-BFS algorithm.…”

Section: Construction Of Regular Coversmentioning

confidence: 99%

“…In this example, there are original 176 edges obtained from Step 1 of the EB approach. Using the fill-in algorithm developed by Huang and Darwiche (1996), we add 84 additional edges in Step 2 to generate a regular cover. The distribution of ∆ρ i,j of all edges (i, j) is shows in Figure 11 as well that of additional edges.…”

Section: Robustness Testsmentioning

confidence: 99%

See 2 more Smart Citations

Robustness to Dependency in Portfolio Optimization Using Overlapping Marginals

Doan

Natarajan

2015

Operations Research

View full text Add to dashboard Cite

In this paper, we develop a distributionally robust portfolio optimization model where the robustness is across different dependency structures among the random losses. For a Fréchet class of discrete distributions with overlapping marginals, we show that the distributionally robust portfolio optimization problem is efficiently solvable with linear programming. To guarantee the existence of a joint multivariate distribution consistent with the overlapping marginal information, we make use of a graph theoretic property known as the running intersection property. Building on this property, we develop a tight linear programming formulation to find the optimal portfolio that minimizes the worst-case Conditional Value-at-Risk measure. Lastly, we use a data-driven approach with financial return data to identify the Fréchet class of distributions satisfying the running intersection property and then optimize the portfolio over this class of distributions. Numerical results in two different datasets show that the distributionally robust portfolio optimization model improves on the sample-based approach

show abstract

Section: Construction Of Regular Coversmentioning

confidence: 99%

Section: Construction Of Regular Coversmentioning

confidence: 99%

See 1 more Smart Citation

Robustness to Dependency in Portfolio Optimization Using Overlapping Marginals

Doan

Natarajan

2015

Operations Research

View full text Add to dashboard Cite

show abstract

“…The availability of such a BN enables the computation of the marginal probability density of a group of variables by conditioning on the rest. This is commonly performed by applying the junction tree algorithm [9]. The result of the latter is a set of probabilities, for all the discrete states of a variable Z, conditioned on the observed evidence e: p k = p(Z = k|e).…”

Section: Inferencementioning

confidence: 99%

Predicting human intention in visual observations of hand/object interactions

Song

Kyriazis

Oikonomidis

et al. 2013

2013 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

Abstract-The main contribution of this paper is a probabilistic method for predicting human manipulation intention from image sequences of human-object interaction. Predicting intention amounts to inferring the imminent manipulation task when human hand is observed to have stably grasped the object. Inference is performed by means of a probabilistic graphical model that encodes object grasping tasks over the 3D state of the observed scene. The 3D state is extracted from RGB-D image sequences by a novel vision-based, markerless hand-object 3D tracking framework. To deal with the high-dimensional statespace and mixed data types (discrete and continuous) involved in grasping tasks, we introduce a generative vector quantization method using mixture models and self-organizing maps. This yields a compact model for encoding of grasping actions, able of handling uncertain and partial sensory data. Experimentation showed that the model trained on simulated data can provide a potent basis for accurate goal-inference with partial and noisy observations of actual real-world demonstrations. We also show a grasp selection process, guided by the inferred human intention, to illustrate the use of the system for goal-directed grasp imitation.

show abstract

“…If it so happens that the missing components are of particular relevance, one may want to compute the posterior probability of the hidden nodes, given the incomplete measurements. In order to compute that, one can use exact inference methods [35] in the context of BNs, or approximate inference methods [36] if the exact solution is intractable.…”

Section: Benefits For Pbnsmentioning

confidence: 99%

Relationships between probabilistic Boolean networks and dynamic Bayesian networks as models of gene regulatory networks

et al. 2006

View full text Add to dashboard Cite

A significant amount of attention has recently been focused on modeling of gene regulatory networks. Two frequently used large-scale modeling frameworks are Bayesian networks (BNs) and Boolean networks, the latter one being a special case of its recent stochastic extension, probabilistic Boolean networks (PBNs). PBN is a promising model class that generalizes the standard rule-based interactions of Boolean networks into the stochastic setting. Dynamic Bayesian networks (DBNs) is a general and versatile model class that is able to represent complex temporal stochastic processes and has also been proposed as a model for gene regulatory systems. In this paper, we concentrate on these two model classes and demonstrate that PBNs and a certain subclass of DBNs can represent the same joint probability distribution over their common variables. The major benefit of introducing the relationships between the models is that it opens up the possibility of applying the standard tools of DBNs to PBNs and vice versa. Hence, the standard learning tools of DBNs can be applied in the context of PBNs, and the inference methods give a natural way of handling the missing values in PBNs which are often present in gene expression measurements. Conversely, the tools for controlling the stationary behavior of the networks, tools for projecting networks onto sub-networks, and efficient learning schemes can be used for DBNs. In other words, the introduced relationships between the models extend the collection of analysis tools for both model classes.

show abstract

Inference in belief networks: A procedural guide

Cited by 313 publications

References 13 publications

Robustness to Dependency in Portfolio Optimization Using Overlapping Marginals

Robustness to Dependency in Portfolio Optimization Using Overlapping Marginals

Predicting human intention in visual observations of hand/object interactions

Relationships between probabilistic Boolean networks and dynamic Bayesian networks as models of gene regulatory networks

Contact Info

Product

Resources

About