Belief Update in Bayesian Networks Using Uncertain Evidence

Pan, Rong; Peng, Yun; Ding, Zhongli

doi:10.1109/ictai.2006.39

Cited by 28 publications

(31 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note that this is an example of hard evidence. Other types of evidence exist, such as soft evidence [46,34] and virtual evidence [40]. Briefly, these types of evidence assign a probability distribution (soft evidence) or likelihood ratios (virtual evidence) to the domain values of a variable.…”

Section: Inferencementioning

confidence: 99%

Exploiting the Probability of Observation for Efficient Bayesian Network Inference

Mousumi

Grant

2012

Advances in Artificial Intelligence

View full text Add to dashboard Cite

It is well-known that the observation of a variable in a Bayesian network can affect the effective connectivity of the network, which in turn affects the efficiency of inference.Unfortunately, the observed variables may not be known until runtime, which limits the amount of compile-time optimization that can be done in this regard. This thesis considers how to improve inference when users know the likelihood of a variable being observed. It demonstrates how these probabilities of observation can be exploited to improve existing heuristics for choosing elimination orderings for inference. Empirical tests over a set of benchmark networks using the Variable Elimination algorithm show reductions of up to 50% and 70% in multiplications and summations, as well as runtime reductions of up to 55%. Similarly, tests using the Elimination Tree algorithm show reductions by as much as 64%, 55%, and 50% in recursive calls, total cache size, and runtime, respectively. iv Acknowledgments I take much pleasure to express my profound gratitude to my supervisor Dr. Kevin Grant for his persistent and inspiring supervision and helping me learn the ABC of Bayesian Networks. Without his endless help and continuous reassurance at the most difficult times, this thesis would not have been a reality. I am indebted to him for his support, encouragement, suggestions, generosity, and the invaluable knowledge he shared with me.

show abstract

Section: Inferencementioning

confidence: 99%

Exploiting the Probability of Observation for Efficient Bayesian Network Inference

Mousumi

Grant

2012

Advances in Artificial Intelligence

View full text Add to dashboard Cite

show abstract

“…Pan et al have previously proposed a similar method for Bayesian networks, which extends the model with virtual evidence nodes whose conditional distributions they iteratively adjust to fit the probability constraints (Alg. 1 in [7]). Indeed, virtually all related work on the subject of soft evidential update has been centred around variations of IPFP [1,5,7,8].…”

Section: Probability Constraints and Iterative Fittingmentioning

confidence: 99%

“…1 in [7]). Indeed, virtually all related work on the subject of soft evidential update has been centred around variations of IPFP [1,5,7,8]. It has even been suggested to directly apply IPFP but reduce the set of variables for which the joint distribution needs to be explicitly updated from X to X E (Alg.…”

Section: Probability Constraints and Iterative Fittingmentioning

confidence: 99%

“…It has even been suggested to directly apply IPFP but reduce the set of variables for which the joint distribution needs to be explicitly updated from X to X E (Alg. 2 in [7]) -by first precomputing the prior P (X E ), applying IPFP to P (X E ) and then running an inference algorithm to update the beliefs on unobserved variables using a model extended with the results of IPFP. Thus, the complexity of every IPFP step is exponential in the size of X E , which renders the algorithm inapplicable as X E grows larger (it is, however, advantageous if X E is very small).…”

Section: Probability Constraints and Iterative Fittingmentioning

confidence: 99%

See 1 more Smart Citation

Soft Evidential Update via Markov Chain Monte Carlo Inference

Jain

Beetz

2010

KI 2010: Advances in Artificial Intelligence

View full text Add to dashboard Cite

Abstract. The key task in probabilistic reasoning is to appropriately update one's beliefs as one obtains new information in the form of evidence. In many application settings, however, the evidence we obtain as input to an inference problem may be uncertain (e.g. owing to unreliable mechanisms with which we obtain the evidence) or may correspond to (soft) degrees of belief rather than hard logical facts. So far, methods for updating beliefs in the light of soft evidence have been centred around the iterative proportional fitting procedure and variations thereof. In this work, we propose a Markov chain Monte Carlo method that allows to directly integrate soft evidence into the inference procedure without generating substantial computational overhead. Within the framework of Markov logic networks, we demonstrate the potential benefit of this method over standard approaches in a series of experiments on synthetic and real-world applications.

show abstract

“…The method used to perform distribute inference is a variation of the Virtual Evidence Method (VEM) algorithm [6]. This method is a simple way to set a desired belief in a target node.…”

Section: Distributed Bayesian Reasoningmentioning

confidence: 99%

A Multi-Agent System with Distributed Bayesian Reasoning for Network Fault Diagnosis

Carrera

Gonzalez-Ordás

García-Algarra

et al. 2011

Advances in Intelligent and Soft Computing

View full text Add to dashboard Cite

In this paper, an innovative approach to perform distributed Bayesian inference using a multi-agent architecture is presented. The final goal is dealing with uncertainty in network diagnosis, but the solution can be of applied in other fields. The validation testbed has been a P2P streaming video service. An assessment of the work is presented, in order to show its advantages when it is compared with traditional manual processes and other previous systems.

show abstract

Belief Update in Bayesian Networks Using Uncertain Evidence

Abstract: Abstract

Cited by 28 publications

References 2 publications

Exploiting the Probability of Observation for Efficient Bayesian Network Inference

Exploiting the Probability of Observation for Efficient Bayesian Network Inference

Soft Evidential Update via Markov Chain Monte Carlo Inference

A Multi-Agent System with Distributed Bayesian Reasoning for Network Fault Diagnosis

Contact Info

Product

Resources

About