Probabilistic Models for Agents' Beliefs and Decisions

TL;DR

This paper introduces a probabilistic epistemic logic (PEL) for reasoning about agents' beliefs, preferences, and behaviors, enabling inference of mental states and unobserved variables from observed actions using Bayesian networks and influence diagrams.

Contribution

It presents a formal semantics for belief statements in PEL and an algorithm for querying these in Bayesian networks, integrating decision modeling with reasoning about mental states.

Findings

PEL provides a formal framework for reasoning about beliefs and preferences.

The algorithm enables querying PEL formulas efficiently in Bayesian networks.

Modeling decision processes as influence diagrams allows inference of unobserved variables.

Abstract

Many applications of intelligent systems require reasoning about the mental states of agents in the domain. We may want to reason about an agent's beliefs, including beliefs about other agents; we may also want to reason about an agent's preferences, and how his beliefs and preferences relate to his behavior. We define a probabilistic epistemic logic (PEL) in which belief statements are given a formal semantics, and provide an algorithm for asserting and querying PEL formulas in Bayesian networks. We then show how to reason about an agent's behavior by modeling his decision process as an influence diagram and assuming that he behaves rationally. PEL can then be used for reasoning from an agent's observed actions to conclusions about other aspects of the domain, including unobserved domain variables and the agent's mental states.