Using Potential Influence Diagrams for Probabilistic Inference and Decision Making

TL;DR

This paper introduces potential influence diagrams as a generalization of conditional influence diagrams, enabling efficient probabilistic inference and decision analysis, and explores their properties and conversions.

Contribution

It presents a detailed analysis of potential influence diagrams, their relationship with conditional influence diagrams, and methods for converting between them.

Findings

Potential influence diagrams enable efficient inference calculations.

Methods for converting potential influence diagrams into conditional influence diagrams.

Insight into the properties and operations of potential influence diagrams.

Abstract

The potential influence diagram is a generalization of the standard "conditional" influence diagram, a directed network representation for probabilistic inference and decision analysis [Ndilikilikesha, 1991]. It allows efficient inference calculations corresponding exactly to those on undirected graphs. In this paper, we explore the relationship between potential and conditional influence diagrams and provide insight into the properties of the potential influence diagram. In particular, we show how to convert a potential influence diagram into a conditional influence diagram, and how to view the potential influence diagram operations in terms of the conditional influence diagram.