Expected Utility Networks

TL;DR

Expected Utility Networks (EUNs) are a new graphical model that simultaneously represents probability and utility dependencies, enabling efficient computation of conditional expected utilities for decision-making in AI and economics.

Contribution

This paper introduces EUNs, a novel graphical framework that unifies probability and utility representations with a new notion of conditional utility independence.

Findings

EUNs use undirected graphs with probability and utility arcs.

Conditional EU independence is characterized by node separation in EUNs.

EUNs facilitate strategic inference through conditional expected utility computation.

Abstract

We introduce a new class of graphical representations, expected utility networks (EUNs), and discuss some of its properties and potential applications to artificial intelligence and economic theory. In EUNs not only probabilities, but also utilities enjoy a modular representation. EUNs are undirected graphs with two types of arc, representing probability and utility dependencies respectively. The representation of utilities is based on a novel notion of conditional utility independence, which we introduce and discuss in the context of other existing proposals. Just as probabilistic inference involves the computation of conditional probabilities, strategic inference involves the computation of conditional expected utilities for alternative plans of action. We define a new notion of conditional expected utility (EU) independence, and show that in EUNs node separation with respect to the probability and utility subgraphs implies conditional EU independence.