PROBABILITY & STATISTICS

A Formal Semantics of Influence in Bayesian Reasoning

Resource type
Authors/contributors
Title
A Formal Semantics of Influence in Bayesian Reasoning
Abstract
This paper proposes a formal definition of influence in Bayesian reasoning, based on the notions of state (as probability distribution), predicate, validity and conditioning. Our approach highlights how conditioning a joint entwined/entangled state with a predicate on one of its components has ‘crossover’ influence on the other components. We use the total variation metric on probability distributions to quantitatively measure such influence. These insights are applied to give a rigorous explanation of the fundamental concept of d-separation in Bayesian networks.
Publication
Schloss Dagstuhl - Leibniz-Zentrum fuer Informatik GmbH, Wadern/Saarbruecken, Germany
Date
2017
Language
en
DOI
10/ggdgbc
Accessed
2019-11-24T12:11:15Z
Library Catalog
DataCite
Extra
ZSCC: 0000012
Citation
Jacobs, B., & Zanasi, F. (2017). A Formal Semantics of Influence in Bayesian Reasoning. Schloss Dagstuhl - Leibniz-Zentrum Fuer Informatik GmbH, Wadern/Saarbruecken, Germany. https://doi.org/10/ggdgbc
Processing time: 0.02 seconds

Graph of references

(from Zotero to Gephi via Zotnet with this script)
Graph of references