Causal Networks: Semantics and Expressiveness

Machine intelligence and pattern recognitionPublished 1 January 1990

Thomas Verma, Judea Pearl

Citations349

TL;DR

This paper shows that the graphical criterion called d-separation is a sound rule for reading independencies from any DAG based on a causal input list drawn from a graphoid and may be extended to cover DAGs that represent functional dependencies as well as conditional dependencies.

Abstract

Dependency knowledge of the form "x is independent of y once z is known" invariably obeys the four graphoid axioms, examples include probabilistic and database dependencies. Often, such knowledge can be represented efficiently with graphical structures such as undirected graphs and directed acyclic graphs (DAGs). In this paper we show that the graphical criterion called d-separation is a sound rule for reading independencies from any DAG based on a causal input list drawn from a graphoid. The rule may be extended to cover DAGs that represent functional dependencies as well as conditional dependencies.

Keywords

Computer Science

Artificial IntelligenceFusion, propagation, and structuring in belief networks

2,152 Citations1986Judea Pearl

Journal of the Royal Statistical Society Series B (Statistical Methodology)Conditional Independence in Statistical Theory

1,546 Citations1979A. P. Dawid

ACM eBooksReverend Bayes on Inference Engines: A Distributed Hierarchical Approach

751 Citations2022Judea Pearl

Generalizations of Bayes likelihood-ratio updating rule are presented which facilitate an asynchronous propagation of the impacts of new beliefs and/or new evidence in hierarchically organized inference structures with multi-hypotheses variables.

ACM Transactions on Database SystemsMultivalued dependencies and a new normal form for relational databases

720 Citations1977Ronald Fagin

It is shown that every relation schema can be decomposed into a family of relation schemata in fourth normal form without loss of information (that is, the original relation can be obtained from the new relations by taking joins).

View PDF

Operations ResearchProbabilistic Inference and Influence Diagrams

498 Citations1988Ross D. Shachter

This work develops algorithms to address questions of inference within a probabilistic model represented as an influence diagram and uses the conditional independence implied by the diagram's structure to determine the information needed to solve a given problem.

Machine intelligence and pattern recognitionA Constraint – Propagation Approach to Probabilistic Reasoning

169 Citations1986Judea Pearl

The paper demonstrates that strict adherence to probability theory does not preclude the use of concurrent, self-activated constraint-propagation mechanisms for managing uncertainty and maintains local records of sources-of-belief.

National Conference on Artificial IntelligenceThe logic of representing dependencies by directed graphs

119 Citations1987Judea Pearl, Thomas Verma

The etiology and prevalence of OH in the general population is outlined, its relationship with morbidity and mortality is summarized, a diagnostic and therapeutic algorithm is proposed, and current challenges and future perspectives are delineated.

Machine intelligence and pattern recognitionIntelligent Probabilistic Inference

82 Citations1986Ross D. Shachter

This chapter focuses on arbitrary conditional probability distributions from a given probabilistic model, where the conditional independence present in the model can be exploited to avoid having to construct or manipulate the full joint distribution.