Reverend Bayes on Inference Engines: A Distributed Hierarchical Approach

@article{Pearl1982ReverendBO,
  title={Reverend Bayes on Inference Engines: A Distributed Hierarchical Approach},
  author={Judea Pearl},
  journal={Probabilistic and Causal Inference},
  year={1982}
}
  • J. Pearl
  • Published 18 August 1982
  • Computer Science
  • Probabilistic and Causal Inference
This paper presents generalizations of Bayes likelihood-ratio updating rule which facilitate an asynchronous propagation of the impacts of new beliefs and/or new evidence in hierarchically organized inference structures with multi-hypotheses variables. The computational scheme proposed specifies a set of belief parameters, communication messages and updating rules which guarantee that the diffusion of updated beliefs is accomplished in a single pass and complies with the tenets of Bayes… 

Figures from this paper

A Computational Model for Causal and Diagnostic Reasoning in Inference Systems

TLDR
A representation of evidential relationships which permits updating of belief in two simultaneous modes: causal and diagnostic is introduced, which extends the hierarchical tree representation by allowing multiple causes to a given manifestation.

Distributed Reasoning and Learning in Bayesian Expert Systems

This paper presents Bayesian networks as a framework for distributed reasoning in expert systems. We discuss methods for evidence propagation, for learning, with emphasis on sequential learning, and

A Constraint-Propagation Approach to Probabilistic Reasoning

An algorithm for the recovery of both target joint beliefs and full belief from Bayesian networks

TLDR
A new algorithm is presented that will combine the two types of algorithms so that one can recover a desired joint belief as well its all the single beliefs.

Convex Combination Belief Propagation Algorithms

TLDR
Modifications to the standard belief propagation algorithms are described that lead to methods that converge to unique solutions on graphical models with arbitrary topology and potential functions.

Inference Networks and the Evaluation of Evidence: Alternative Analyses

TLDR
The use of these three methods for generating and analyzing probabilistic inference networks is illustrated in an analysis of a mass of evidence in a celebrated American law case.

Study of the Bayesian networks

  • Yonghui Cao
  • Computer Science
    2010 International Conference on E-Health Networking Digital Ecosystems and Technologies (EDT)
  • 2010
TLDR
The chain rule theorem is introduced to do the necessary calculations in Bayesian networks and the influence diagrams method to convert beliefs of an agent into actions is introduced.

Efficient Gradient-Based Inference through Transformations between Bayes Nets and Neural Nets

TLDR
The choice of parameterization greatly influences the efficiency of gradient-based posterior inference; it is shown that they are often complementary to eachother, it is clarified when each parameterization is preferred and how inference can be made robust.

7 Inference in Bayesian Networks : A Historical Perspective

TLDR
A historical perspective is provided on the key role that Judea Pearl has played in initiating and inspiring many of the technical developments that have formed and continue to form the basis of work in this area of Bayesian networks.

Learning Link-Probabilities in Causal Trees

TLDR
A learning algorithm is presented which given the structure of a causal tree, will estimate its link probabilities by sequential measurements on the leaves only, and remains robust to measurement imprecisions.
...

References

SHOWING 1-10 OF 11 REFERENCES

Subjective Bayesian methods for rule-based inference systems

TLDR
A subjective Bayesian inference method that realizes some of the advantages of both formal and informal approaches, and modifications needed to deal with the inconsistencies usually found in collections of subjective statements are described.

On the Independence Assumption Underlying Subjective Bayesian Updating

A Retrospective View of the Hearsay-II Architecture

TLDR
Experiences gained while successfully applying the Hearsay-II architecture to the problem of speech understanding are described, and the paradigm of viewing problem solving in terms of hypothesize-and-test actions distributed among distinct representations of the problem has been shown to be computationally feasible.

Empirical Studies of Cascaded Inference in Jurisprudence: Methodological Consideration

  • Empirical Studies of Cascaded Inference in Jurisprudence: Methodological Consideration

Psychology Research Report

  • Psychology Research Report
  • 1980

DDI Handbook for Decision Analysis, Decision and Design Inc

  • DDI Handbook for Decision Analysis, Decision and Design Inc
  • 1973

Assuming now that P(A) is valid, let us examine the validity of q(B), where B is any child of A. By definition (Equation (11.6)), q(B) should satisfy: q(B i ) = P(B i | D u (B)) = ∑ P(B i | A j )P

    Chapter 11 Reverend Bayes on Inference Engines: A Distributed Hierarchical Approach

      Belief Propagation in Hierarchical Inference Structures

      • Belief Propagation in Hierarchical Inference Structures
      • 1982