Agent belief autonomy in open multi-agent systems

K. S. Barber, J. Park

Research output: Chapter in Book/Report/Conference proceedingChapter

5 Citations (Scopus)

Abstract

The goals and the beliefs of an agent are not independent of each other. In order for an agent to be autonomous an agent must control its beliefs as well as tasks/goals. The agent's beliefs about itself, others and the environment are based on its derived models of perceived and communicated information. The degree of an agent's belief autonomy is its degree of dependence on others to build its belief models. We propose source trustworthiness, coverage and cost as factors an agent should use to determine on which sources to rely. Trustworthiness represents how reliable other agents are. Coverage is a measure of an information source's contribution to an agent's information needs. Cost of getting information from a source is defined by the timeliness of information delivery from the source. Since a respective agent only knows about a limited amount of information sources which can communicate with the agent, it must also rely on other agents to share their models describing sources they know about. These factors along with the degree to which a respective agent shares its knowledge about its neighbors are represented in this paper and proposed as contributions to agent's decisions regarding its belief autonomy for respective goals.
Original languageEnglish
Title of host publicationAgents and Computational Autonomy
Subtitle of host publicationPotential, Risks, and Solutions
EditorsM Nickles, M Rovatsos, G Weiss
Place of PublicationBerlin, Germany
PublisherSpringer
Pages7-16
Number of pages10
ISBN (Print)9783540224778
Publication statusPublished - 2004

Publication series

NameLecture Notes in Computer Science
Volume2969

Fingerprint Dive into the research topics of 'Agent belief autonomy in open multi-agent systems'. Together they form a unique fingerprint.

  • Cite this

    Barber, K. S., & Park, J. (2004). Agent belief autonomy in open multi-agent systems. In M. Nickles, M. Rovatsos, & G. Weiss (Eds.), Agents and Computational Autonomy: Potential, Risks, and Solutions (pp. 7-16). (Lecture Notes in Computer Science; Vol. 2969). Springer.