Multiagent Distributed Interpretation With Multiply Sectioned Bayesian Netowrks

8/27/98


Click here to start


Table of Contents

Multiagent Distributed Interpretation With Multiply Sectioned Bayesian Netowrks

Recent Advances

Common approaches in MAS

MAS Area of Focus

Recent Advance

Foundations

Background

Background

*Bayesian Networks (BN)

A Trivial Example BN [L & S 88]

Multiply Sectioned Bayesian Networks (MSBNs)

*The d-sepset: Interface b/w Subnets in a MSBN

*Hypertree MSDAG: Top Level Structure of a MSBN

*Compilation of a MSBN into a Linked Junction Forest

*Inference in a Single-agent Oriented MSBN

What Can be Gained y Using a MSBN?

Why Using MSBNs for Distributed Interpretation?

Major Issues in Extending MSBNs to MASs

What is the semantics of a subnet?

What is the semantics of the jpd in a MSBN?

What is the semantics of the jpd in a MSBN?

What is the semantics of the jpd in a MSBN?

How do we ensure coherent inference?

How do we ensure coherent inference?

How do we ensure coherent inference?

How do we ensure coherent inference?

How do we ensure coherent inference?

How to build a MSBN by multiple developers?

Global structure vs each agent's know-how

Global structure vs each agent's know-how

What if agents are not organized into a MSBN?

Prospects

Prospects for distributed interpretation

Prospects for distributed interpretation

Author: Y. Xiang

Email: yxiang@cs.uregina.ca

Home Page: http://cs.uregina.ca/~yxiang