Title: An Introduction to Bayesian Networks for Multi-Agent Systems
1An Introduction to Bayesian Networksfor
Multi-Agent Systems
By Vijay Sargunar.M.M
2A Bayesian Network
Battery
Gas
Radio
Ignition
Starts
Moves
- Features of a Cars Electrical System and Engine
3Types of Bayesian Networks Trouble shooting
4Types of Bayesian Networks Diagnosis
5Guessing the state of the problem domain
- The agent makes observations on the domain,
guesses the state of the problem domain based on
the observations and its prior knowledge about
the domain, and determines the most appropriate
action based on its belief and goal. - Agents constructed from a rule-based system uses
a symbolic knowledge representation. - We consider agents using symbolic knowledge
representations and reasoning explicitly about
the state of the domain.
6Reasons for guessing
- Agent does not observe some aspects of the domain
estimated indirectly through observable - Relations between domain events are often
uncertain - Observations themselves may be imprecise,
ambiguous, vague, noisy, and unreliable - Lack of resources to observe all incomplete
information - Event relations are certain Impractical to
analyze all of them explicitly.
7Bayesian Networks for Probabilistic
Reasoning
- BN is used as a concise graphical repn. of a
decision makers probabilistic knowledge of an
uncertain domain. - BN is primarily used to update the belief of an
agent from that of a prior belief to a posterior
belief when evidence is received. - Probabilistic reasoning using Bayesian networks
is called belief updating.
8A Simple Digital Circuit
- Agent monitoring a digital circuit.
r
a
e
d
t
c
b
g
Agents prior belief P(a,b,c,d,e,g,r,t) P(a0,b0,
c0,d0,e1,gnormal,rnormal,tnormal)
0.2 P(a0,b0,c0,d0,e0,gnormal,rnormal,tabn
ormal) 0.009
9Local Computation and Message Passing
habit
Posterior distribution P(h/ty) is to be
computed. Node t send message to c. P(ty/c)
0.85,0.05 Node c sends its message to node
h. P(ty/h) ?P(ty,c/h) ?P(ty/c,h)P(c/h)
?P(ty/c)P(c/h) 0.13,0.69 When h receives
P(ty/h) it can compute P(h/ty) const
P(ty/h)P(h) 0.3054,0.6946
cavity
toothache
10Junction Tree Representation
Q
S
C
c
t,c
c,h
P(t/c)
P(c/h)P(h)
11Multi-Agent Uncertain Reasoning with Multiply
Sectioned Bayesian Networks
- MSBN is knowledge representation formalism for
multi-agent uncertain reasoning. - Areas used
- Aircraft (Intricate Machines)
- Monitoring Trouble shooting in Chemical
processes. - Problem domain spread over a large
geographical area.
12 A digital system consists of five components
(U0,..,U4) from different vendors. Each vendor
has built an agent capable of monitoring the
component. Here, we assume identical faulty
behavior of gates only for convenience.
A digital system.
13The core representation of each agent is a subnet
(Di, I0,,4).The internal structure and
parameter of each subnet is unknown to other
vendors.
14Multiply Sectioned Bayesian Networks (MSBNs)
- A set of Bayesian subnets that collectively
define a BN. - Interface subnets renders them conditionally
independent. - Compiled into a linked junction forest (LJF) for
inference. - In a single-agent MSBN, evidence is entered one
subnet at a time. - In a multi-agent MSBN, evidence are entered
asynchronously at multiple subnets in parallel.
15Distributed Multi-agent Inference
- Pass messages among agents effectively so that
each agent can update its belief correctly with
respect to the observations made by all agents in
the system.
16Model Construction Verification
- Integrate an MSBN-based Multi-agent system from
agents developed by independent vendors. - Verification Process becomes subtle when agents
are built by independent vendors and vendors
know-how needs to be protected Multi-agent
distributed verification.
17Software Tool
18Research Areas Conclusion.
- Identification of Trustworthiness (Information
quality) assessment of other agents - If Prior beliefs quality of Information is less
or excessively high the entire network gets
distorted towards underestimation or
overestimation of other agents. - Generation of Plans by an agent based on a
Bayesian networks situation assessment.
19Thank You
20(No Transcript)
21(No Transcript)
22(No Transcript)
23(No Transcript)