73
1 An Implementation of An Implementation of Multiply Sectioned Multiply Sectioned Bayesian Networks Bayesian Networks Metron, Inc. Chris Boner Thor Whalen

An Implementation of Multiply Sectioned Bayesian Networks

  • Upload
    mirit

  • View
    40

  • Download
    0

Embed Size (px)

DESCRIPTION

An Implementation of Multiply Sectioned Bayesian Networks. Metron, Inc. Chris Boner Thor Whalen. Outline. Multiply sectioned Bayes nets (MSBN) Problem formulation and elements of a solution Using a junction tree to construct an MSBN Matlab tool. Multiply Sectioned Bayes Net. - PowerPoint PPT Presentation

Citation preview

Page 1: An Implementation of Multiply Sectioned Bayesian Networks

1

An Implementation of Multiply An Implementation of Multiply Sectioned Bayesian NetworksSectioned Bayesian Networks

Metron, Inc.

Chris Boner

Thor Whalen

Page 2: An Implementation of Multiply Sectioned Bayesian Networks

2

OutlineOutline

• Multiply sectioned Bayes nets (MSBN)

• Problem formulation and elements of a solution

• Using a junction tree to construct an MSBN

• Matlab tool

Page 3: An Implementation of Multiply Sectioned Bayesian Networks

3

Multiply Sectioned Bayes NetMultiply Sectioned Bayes Net

• What is a Bayes Net?• What is a Multiply Sectioned Bayes Net (MSBN)?• Motivation

Page 4: An Implementation of Multiply Sectioned Bayesian Networks

4

What is a Bayes Net?What is a Bayes Net?• A Bayes Net is a representation of

a probability distribution P(V) on a set V=X1, ..., Xn of variables

Page 5: An Implementation of Multiply Sectioned Bayesian Networks

5

What is a Bayes Net?What is a Bayes Net?• A Bayes Net is a representation of

a probability distribution P(V) on a set V=X1, ..., Xn of variables

• A BN consists of – A Directed Acyclic Graph (DAG)

• Nodes: Variables of V

• Edges: Causal relations

A DAG is a directed graph with no directed cycles

The above directed graph is a DAG

Now this graph IS NOT a DAG because it has a directed cycle

Directed cycle

X1 X2 X3

X4 X5

X6

X7X8

X9 X10

X11 X12 X13

Page 6: An Implementation of Multiply Sectioned Bayesian Networks

6

What is a Bayes Net?What is a Bayes Net?• A Bayes Net is a representation of

a probability distribution P(V) on a set V=X1, ..., Xn of variables

• A BN consists of – A Directed Acyclic Graph (DAG)

• Nodes: Variables of V

• Edges: Causal relations

– A list of conditional probability distributions (CPDs); one for every node of the DAG

X1 X2 X3

X4 X5

X6

X7X8

X9 X10

X11 X12 X13

Etc...

Page 7: An Implementation of Multiply Sectioned Bayesian Networks

7

- i.e. P(A , B | C) = P(A | C) P(B | C)

What is a Bayes Net?What is a Bayes Net?• A Bayes Net is a representation of

a probability distribution P(V) on a set V=X1, ..., Xn of variables

• A BN consists of – A Directed Acyclic Graph (DAG)

• Nodes: variables of V

• Edges: Causal relations

– A list of conditional probability distributions (CPDs); one for every node of the DAG

• The DAG characterizes the (in)dependence structure of P(V)

X1 X2 X3

X4 X5

X6

X7X8

X9 X10

X11 X12 X13

A C

BA and are independent given

B

C

- i.e. P(A | B, C) = P(A | C)

We will say that C separates A and B

Page 8: An Implementation of Multiply Sectioned Bayesian Networks

8

What is a Bayes Net?What is a Bayes Net?• A Bayes Net is a representation of

a probability distribution P(V) on a set V=X1, ..., Xn of variables

• A BN consists of – A Directed Acyclic Graph (DAG)

• Nodes: variables of V

• Edges: Causal relations

– A list of conditional probability distributions (CPDs); one for every node of the DAG

• The DAG characterizes the (in)dependency structure of P(V)

• The CPDs characterize the probabilistic and/or deterministic relations between parent states and children states

X1 X2 X3

X4 X5

X6

X7X8

X9 X10

X11 X12 X13

Page 9: An Implementation of Multiply Sectioned Bayesian Networks

9

X7

X3X1 X2

X5

X8

X4

X11 X13

X6

X7

X12Evidence

What is a Bayes Net?What is a Bayes Net?

• The prior distributions on the variables of parentless nodes, along with the CPDs of the BN, induce prior distribution—called “beliefs” in the literature—on all the variables

• If the system receives evidence on a variable: – this evidence impacts its belief,– along with the beliefs of all other

variables

X9 X10

Parentless nodes

Page 10: An Implementation of Multiply Sectioned Bayesian Networks

10

• Subnets of the BN are maintained independently

• Each subnet locally integrates evidence it receives

• When inter-subnets communication is possible, messages are passed that enable fusion of evidence received by other subnets

Evidence

Evidence

What is a Multiply Sectioned Bayes Net?What is a Multiply Sectioned Bayes Net?

Page 11: An Implementation of Multiply Sectioned Bayesian Networks

11

• Multi-agent systems where:– each agent only has partial knowledge of the

domain– communication among agents is limited

• some agent-agent connections may be impossible, sporadic and/or low bandwidth

– decisions must be made by agents based on local observations and limited information from other agents

MotivationMotivation

Page 12: An Implementation of Multiply Sectioned Bayesian Networks

12

Motivation (2)Motivation (2)

• Distributed computing for reusable systems where:– probabilistic knowledge can be captured once

and used for multiple cases– queries and evidence will be localized, that is,

there are phases when• new evidence and queries are repeatedly directed

to small parts of network• only a small part of the network is needed for

decision-making

Page 13: An Implementation of Multiply Sectioned Bayesian Networks

13

• Problem specification

• A naïve solution

• A less naïve solution

• Sufficient information

• Communication graph considerations

Problem Formulation Problem Formulation and elements of a solutionand elements of a solution

Page 14: An Implementation of Multiply Sectioned Bayesian Networks

14

Problem SpecificationProblem SpecificationGiven: • A BN on V={X1, ..., Xn}

• A number of agents, each having:– Qi: a set of query variables

– Ei: a set of evidence variables

Page 15: An Implementation of Multiply Sectioned Bayesian Networks

15

Problem SpecificationProblem SpecificationGiven: • A BN on V={X1, ..., Xn}

• A number of agents, each having:– Qi: a set of query variables

– Ei: a set of evidence variables

Determine: • An agent communication graph

• A subset Si of V for each agent

• An inference protocol that specifies – How to fuse evidence and messages

received from other agents– The content of messages between

agents

Page 16: An Implementation of Multiply Sectioned Bayesian Networks

16

Every agent has a copy of the entire Bayes net

Evidence

Agents communicate evidence (findings or likelihood functions) that are re-propagated through each copy of the BN

A Naïve SolutionA Naïve Solution

Page 17: An Implementation of Multiply Sectioned Bayesian Networks

17

Pros• Each agent’s queries

are as informed as possible once all the evidence it has received is propagated

• Inter-agent communications require relatively low bandwidth

Cons• Could be a colossal

waste of memory and processing time

• Each agent may be able to achieve fully informed queries by representing a much smaller section of BN

A Naïve SolutionA Naïve Solution

Page 18: An Implementation of Multiply Sectioned Bayesian Networks

18

• The previous solution allows each agent to compute the posterior prob. of all the variables

• But all the agent is interested in is the posterior of its query variables

• Hence it is sufficient for every agent to only represent

- its query variables, - its evidence variables, - the evidence variables

of the other agents• Contra: Could be a

colossal waste of memory and processing time

A Less Naïve SolutionA Less Naïve Solution

Query

variables

Query

variables

Query

variables

Query

variables

Evidence

variables

Evidence

variables

Evidence

variables

Evidence

variables

Page 19: An Implementation of Multiply Sectioned Bayesian Networks

19

Sufficient informationSufficient information

B

CD

A

E

F

GH

I

JK

L

M

Evidence variables

Query variables

Agent 1

Agent 2

Agent 3

Agent 4

Specifications

• A Bayes net

• A number of agents, each having

- query variables

- evidence variables

Page 20: An Implementation of Multiply Sectioned Bayesian Networks

20

Sufficient informationSufficient information

B

CD

A

E

F

GH

I

JK

L

M

Agent 1

E F G

H I J

K L M

E F G

H I J

K L M

A B

E F G K L MH I J

Specifications

• A Bayes net

• A number of agents, each having

- query variables

- evidence variables

Agent 2

Agent 3

Agent 4

The naïve solution

• Agents contain their own query and evidence variables

• In order to receive evidence from the other agents, agent 1 must represent variables E, F, G, H, I, J, K, L, and M

Agent 1

Agent 2 Agent 3 Agent 4

Page 21: An Implementation of Multiply Sectioned Bayesian Networks

21

E F G K L MH I J

Sufficient informationSufficient information

B

CD

A

E

F

GH

I

JK

L

M

Agent 1

E F G K L MH I J

A B

Specifications

• A Bayes net

• A number of agents, each having

- query variables

- evidence variables

Agent 2

Agent 3

Agent 4

The naïve solution

• Agents contain their own query and evidence variables

• In order to receive evidence from the other agents, agent 1 must represent variables E, F, G, H, I, J, K, L, and M

Y

ZX

separatesZ and X Y

Note that

whether Y is equal to:

• {K,L,M},

• {H,J,I}, or

• {E,F,G}.YY

Agent 1 must

represent many

variables!

How else could

the other agents

communicate their

evidence?

Page 22: An Implementation of Multiply Sectioned Bayesian Networks

22

P(X,Z|eY)

Sufficient informationSufficient information

B

C D

A

E

F

G

separatesZ and X Y

Z

X

Y

X = {A,B}

Z = {C,D}

Z = {C,D}

Y = {E,G,F}

P(Y|Z) = P(Y|X,Z)

→→

Likelihood given X and Z

of evidence on Y

Likelihood given Z

of evidence on Y=

→ It is sufficient for agent 2 to send its posterior on Z

to agent 1 for the latter to compute its posterior on X

Agent 1

Agent 2

P(Y,Z|eY)

P(X,Z)

P(Z|eY)

P(Y,Z)

ΣY

x P(X,Z)P(Z)-1

evidence eY

Page 23: An Implementation of Multiply Sectioned Bayesian Networks

23

P(X,Z|eY)

Sufficient informationSufficient information

B

C D

A

E

F

G

separatesZ and X Y

Z

X

Y

X = {A,B}

Z = {C,D}

Z = {C,D}

Y = {E,G,F}

P(Y|X,Z) = P(Y|Z)

→→

Likelihood given X and Z

of evidence on Y

Likelihood given Z

of evidence on Y=

→ It is sufficient for agent 2 to send its posterior on Z

to agent 1 for the latter to compute its posterior on X

Agent 1

Agent 2

P(Y,Z|eY)

P(Z|eY)

P(eY)-1

P(Z)-1P(X,Z) P(Z|eY)

P(Z)-1= P(X,Z) P(Z,eY)

P(eY)-1= P(X,Z) P(eY|Z)

P(eY)-1= P(X,Z) P(eY|X,Z)

P(X,Z|eY)P(eY)-1= P(X,Z,eY) =

Because:

P(eY|Z)P(eY|X,Z)

Page 24: An Implementation of Multiply Sectioned Bayesian Networks

24

E F G K L MH I J

Sufficient informationSufficient information

E F G K L MH I J

A B

Specifications

• A Bayes net

• A number of agents, each having

- query variables

- evidence variables

The naïve solution

• Agents contain their own query and evidence variables

• In order to receive evidence from the other agents, agent 1 must represent variables E, F, G, H, I, J, K, L, and M

C D

E F G K L MH I J

A B

C D C D C D

Using separation

• Agent 1 only needs to represent two extra variables

• Agent 1 may compute its posterior queries faster from CD than from EFGHIJK

• Communication lines need to transmit two variables instead of three

E F G K L MH I JE F G K L MH I J

C D C D C DC D C D C D

Page 25: An Implementation of Multiply Sectioned Bayesian Networks

25

Communication Graph ConsiderationsCommunication Graph Considerations

1

2

3

4

5

6

Agent 6 receives info from agent 1 through both agent 4 and 5.

How should subnet 6 deal with possible redundancy?

A communication graphOne solution (often adopted) would be to impose a tree structure to the communication graph

Page 26: An Implementation of Multiply Sectioned Bayesian Networks

26

Communication Graph ConsiderationsCommunication Graph Considerations

• When choosing the communication graph, one should take into consideration

- The quality of the possible communication lines- The processing speed of the agents- The importance of given queries

If this is the key decision-making agent

...then this communication graph is more appropriate… than this one

Page 27: An Implementation of Multiply Sectioned Bayesian Networks

27

• In a tree communication graph every edge is the only communication line

between two parts of the network• Hence it must deliver enough information so that the evidence received in

one part may convey its impact to the query variables of the other part• We restrict ourselves to the case where every node represented by an agent can be queried or receive evidence• In this case it is sufficient that the set of variables Z, that will be represented in any communication line, separates the set X of variables of one side of the network from the set Y of variables of the other side

Communication Graph ConsiderationsCommunication Graph Considerations

Z

YX

Page 28: An Implementation of Multiply Sectioned Bayesian Networks

28

Using a Junction Tree Using a Junction Tree to construct an MSBNto construct an MSBN

• The junction tree and its use• Building a Junction tree

– Moralization– Triangulation– The junction graph– From junction graph to junction tree

• Partitioning the junction tree• Adding and removing agents• A note on continuous variables

Page 29: An Implementation of Multiply Sectioned Bayesian Networks

29

a

b c

d e

f

g

h

Secondary Structure/Junction Tree• multi-dim. random variables• joint probabilities (potentials)

Bayesian Network• one-dim. random variables• conditional probabilities

abd

ade

ace

ceg

eghdef

ad ae ce

de eg

The junction tree and its useThe junction tree and its use

Page 30: An Implementation of Multiply Sectioned Bayesian Networks

30

abd

ade

ace

ceg

eghdef

• A junction tree is a graphical model of a probability space:

- Nodes of a JT are sets of variables

- Edges of a JT (called sepsets) are labeled by the intersection of the set of variables of the nodes they join

• The set of variables Z of any edge of a JT separates the set of variables of the sub-trees of both sides of this edge

ad

de

ae ce

eg

e.g.{a,e} separates {a,b,d,e,f} and {a,c,e,g,h}

a

b c

d e

f

g

h

The junction tree and its useThe junction tree and its use

Page 31: An Implementation of Multiply Sectioned Bayesian Networks

31

abd

ade

ace

ceg

eghdef

ad

de

ae ce

eg

The junction tree and its useThe junction tree and its use

So any partition a junction tree into sub-trees will allow for distributed inference

a

b

d e

f

a

c

e

g

e

g

h

a,e

e,g

Page 32: An Implementation of Multiply Sectioned Bayesian Networks

32

TopCat

Feat 3Feat 7 Feat 8

Feat 5Feat 4

Feat 6

Feat 1 Feat 2

Sens1b

Sens1a

Sens2b

Sens2a

Example of the Junction Tree ApproachExample of the Junction Tree Approach

Agent 1

Agent 2

Agent 2

Agent 4

query nodes

query nodes

evidence nodes

query nodesevidence nodes

query nodes

evidence nodes

Page 33: An Implementation of Multiply Sectioned Bayesian Networks

33

Building a Junction TreeBuilding a Junction Tree

DAG

Moral Graph

Triangulated Graph

Junction Tree

Identifying Cliques

a

b c

d e

f

g

h a

b c

d e

f

g

ha

b c

d e

f

g

h a

b

d

a

c

ed e

f

a

d e

c

e

g

e

g

h

Page 34: An Implementation of Multiply Sectioned Bayesian Networks

34

TopCat

Feat 3Feat 7 Feat 8

Feat 5Feat 4

Feat 6

Feat 1 Feat 2

Sens1b

Sens1a

Sens2b

Sens2a

Building a Junction TreeBuilding a Junction Tree: MoralizationMoralization

1) Add an edge between every node having a common child.

2) Drop the directions of all other edges.

Page 35: An Implementation of Multiply Sectioned Bayesian Networks

35

TopCat

Feat 3Feat 7 Feat 8

Feat 5Feat 4

Feat 6

Feat 1 Feat 2

Sens1b

Sens1a

Sens1b

Sens1a

Building a Junction TreeBuilding a Junction Tree: TriangulationTriangulation

Add edges to the graph to triangulate (induced) cycles of length greater than three.

This is the only

induced cycle

of length greater

than three

There are only two ways

to triangulate it...So we’ll choose this

way for our example

This way can be shown

to be problematic

Page 36: An Implementation of Multiply Sectioned Bayesian Networks

36

Building a Junction TreeBuilding a Junction Tree: Junction GraphJunction Graph

TopCat

Feat 3Feat 7 Feat 8

Feat 5Feat 4

Feat 6

Feat 1 Feat 2

Sens1b

Sens1a

Sens2b

Sens2a

• A complete subgraph is one with edges between every vertex of the subgraph.

• A clique is a complete subgraph contained in no other complete subgraph.

This is a clique.This is a clique.

This is NOT a clique.Now it IS a clique.

Page 37: An Implementation of Multiply Sectioned Bayesian Networks

37

Building a Junction TreeBuilding a Junction Tree: Junction GraphJunction Graph

TopCat

Feat 3

Feat 7Feat 8

Feat 4

Feat 6

Feat 1 Feat 2

Sens1b

Sens1a

Sens2b

1) Identify Cliques.Every clique corresponds to a node in the JG.

3) Label edges of JG with intersection of cliques.

Feat 1 Feat 2Sens1aSens1b

Feat 5 Feat 6Sens2aSens2b

Feat 1Feat 2TopCat

Feat 5Feat 4Feat 3

Feat 4Feat 7

TopCat

Feat 4Feat 3TopCat Feat 8

Feat 7TopCat

Feat 4

Feat 8Feat 7

2) Draw an edge between two nodes if they share variables.

Feat 1Feat 2

TopCatTopCat

TopCat

Feat 8Feat 7

Feat 4

Feat 3Feat 4

Feat 4

Feat 4

Feat 4TopCat

Feat

7

Feat 5

Feat 7

Feat 5

Sens2a

TopC

at

Page 38: An Implementation of Multiply Sectioned Bayesian Networks

38

Feat 1 Feat 2

Sens1a Sens1b Feat 5 Feat 6

Sens2a Sens2b

Feat 1

Feat 2

TopCat

Feat 5

Feat 4

Feat 3

Feat 4

Feat 7

TopCat

Feat 4

Feat 3

TopCat Feat 8

Feat 7

TopCat

Feat 4

Feat 8

Feat 7

Feat 4

From Junction Graph to Junction TreeFrom Junction Graph to Junction Tree

1

1

1

21

2

1

2

1

2

1) Weight every edge with the number of variables it is labeled with

2) Find a maximal weight spanning (i.e. covering all JG nodes) tree

3) The corresponding subgraph of the JG is a Junction Tree (JT)

Feat 2

TopCatTopCat

TopCat

Feat 8

Feat 4

TopCat

Feat 5

2

2

Feat 1

Feat 4

Feat 4

Feat 7

Feat 4

Feat 7

Feat 7TopCat

Feat 3

The cliques are

the nodes of the JT

The edges of the JT

are called sepsets

Page 39: An Implementation of Multiply Sectioned Bayesian Networks

39

Feat 1 Feat 2

Sens1a Sens1b

Feat 5 Feat 6

Sens2a Sens2b

Feat 1

Feat 2

TopCat

Feat 5

Feat 4

Feat 3

Feat 4

Feat 7

TopCat

Feat 4

Feat 3

TopCat

Feat 8

Feat 7

TopCat

Feat 4

Feat 8

Feat 7

TopCatTopCat

Feat 4

Feat 4

TopCat

Partitioning the Junction TreePartitioning the Junction Tree

• Partition the nodes of the JT:

Sets of the partition → subnets

• Edges between two nodes of different subnets → communication lines

• Desirable for portion of JT inside a subnet to be connected

Feat 1

Feat 2Feat 5

Feat 8

Feat 7

Feat 3

Page 40: An Implementation of Multiply Sectioned Bayesian Networks

40

Corresponding SubnetsCorresponding Subnets

TopCat

Feat 3

Feat 7 Feat 8Feat 5

Feat 4

Feat 1 Feat 2

Feat 1 Feat 2 Feat 5

Feat 6

Feat 7 Feat 8

Feat 1 Feat 2 Feat 5

Feat 7 Feat 8

Evidence EvidenceEvidence

TopCat

Feat 3

Feat 7 Feat 8Feat 5

Feat 4

Feat 1 Feat 2

Sens1b

Sens1a Feat 6

Sens2b

Sens2a

Feat 4

Partitioning the Junction TreePartitioning the Junction Tree

Page 41: An Implementation of Multiply Sectioned Bayesian Networks

41

Adding and removing AgentsAdding and removing Agents

TopCat

Feat 3Feat 7 Feat 8

Feat 5Feat 4

Feat 6

Feat 1 Feat 2

Sens1b

Sens1a

Sens2b

Sens2a

Sens3b

Sens3a

Sens5b

Sens5a

Sens4b

Sens4a

Sens6b

Sens6a

• Here we address the problem of adding and removing agents to the network

• Consider the BN given earlier

• Is it possible to add and remove agents (containing sensor variables) to the

network and perform inference without reconfiguring the network

Page 42: An Implementation of Multiply Sectioned Bayesian Networks

42

Adding and removing Agents (2)Adding and removing Agents (2)

TopCat

Feat 1 Feat 2

Sens1b

Sens1a

Sens3b

Sens3a

Sens5b

Sens5a

• Adding the cliques containing the new variables does not change the structure of the Junction tree, so new agents containing these variables may easily be added and removed, along with a single communication line to the rest of the network.

Etc ...

Feat 1 Feat 2

Sens3a Sens3b

Feat 1

Feat 2

TopCat

Feat 1 Feat 2

Sens5a Sens5b

Feat 1 Feat 2

Sens1a Sens1b

Feat 2

Feat 1

Feat 2

Feat 1

Feat 2

Feat 1

Etc ...

Page 43: An Implementation of Multiply Sectioned Bayesian Networks

43Feat 5

Feat 6

Feat 5

Feat 6

Adding and removing Agents (3)Adding and removing Agents (3)

• Adding the cliques containing the new variables does not change the structure of the Junction tree, so new agents containing these variables may easily be added and removed, along with a single communication line to the rest of the network.

TopCat

Feat 3

Feat 7 Feat 8

Feat 5

Feat 4

Feat 6Sens2b

Sens2a

Sens6b

Sens6a

Sens4b

Sens4a

Etc ...

Feat 5 Feat 6

Sens2a Sens2b

Feat 5

Feat 4

Feat 3

Feat 4

Feat 7

TopCat

Feat 4

Feat 3

TopCat

Feat 8

Feat 7

TopCat

Feat 4

TopCat

Feat 5

Feat 4Feat 7

TopCatFeat 3

Feat 5 Feat 6

Sens4a Sens4b

Feat 5 Feat 6

Sens6a Sens6b

Etc ...

Page 44: An Implementation of Multiply Sectioned Bayesian Networks

44Feat 5

Feat 6

Feat 5

Feat 6

Adding and removing Agents (4)Adding and removing Agents (4)

• It is not desirable to have sensors chained as such since evidence received in one agent must pass through other agents to reach the central agent

• It would be preferable to have the sensor agents communicate with the central agent directly

TopCat

Feat 3

Feat 7 Feat 8

Feat 5

Feat 4

Feat 6Sens2b

Sens2a

Sens6b

Sens6a

Sens4b

Sens4a

Etc ...

Feat 5 Feat 6

Sens2a Sens2b

Feat 5

Feat 4

Feat 3

Feat 4

Feat 7

TopCat

Feat 4

Feat 3

TopCat

Feat 8

Feat 7

TopCat

Feat 4

TopCat

Feat 5

Feat 4Feat 7

TopCatFeat 3

Feat 5 Feat 6

Sens4a Sens4b

Feat 5 Feat 6

Sens6a Sens6b

Etc ...

Page 45: An Implementation of Multiply Sectioned Bayesian Networks

45

Adding and removing Agents (3)Adding and removing Agents (3)

TopCat

Feat 3

Feat 7 Feat 8

Feat 5

Feat 4

Feat 6Sens2b

Sens2a

Sens6b

Sens6a

Sens4b

Sens4a

Etc ...

Feat 5 Feat 6

Sens2a Sens2b

Feat 5

Feat 4

Feat 3

Feat 4

Feat 7

TopCat

Feat 4

Feat 3

TopCat

Feat 8

Feat 7

TopCat

Feat 4

TopCat

Feat 5

Feat 4Feat 7

TopCatFeat 3

Feat 5 Feat 6

Sens4a Sens4b

Feat 5 Feat 6

Sens6a Sens6b

Etc ...

• By adding an extra variable the appropriate clique, we now have a different junction tree structure more fit for our application

Feat 5Feat 6

Feat 5Feat 6

Feat 6

Feat 6

Page 46: An Implementation of Multiply Sectioned Bayesian Networks

46

A note on continuous variablesA note on continuous variables

• Wish to extend JT inference to handle continuous variables.

• JT inference ↔ Potential manipulation• In general, a potential Φ on X=X1, ..., Xn is

a function Φ: X1 x ... x Xn → [0,+∞)• We need to define

– Multiplication (and division) of two potentials• Use function multiplication for this

– marginalization of a potential• use integration instead of summation

Page 47: An Implementation of Multiply Sectioned Bayesian Networks

47

• Prior distributions and all possible evidence likelihood functions must be represented algebraically by a class of functions closed under multiplication and integration

• If the class of functions doesn’t encompass the prior distributions and evidence exactly, the question arises whether approximate inference or discretization might yield better results

A note on continuous variablesA note on continuous variables

Page 48: An Implementation of Multiply Sectioned Bayesian Networks

48

• Algebraic manipulation is not a straightforward computational task

• Exact integration is not always possible

• Is numerical integration necessarily better than approximate inference or discretization?

A note on continuous variablesA note on continuous variables

Page 49: An Implementation of Multiply Sectioned Bayesian Networks

49

MatLab ToolMatLab Tool

• Setup

• Functionality

• Adding and removing subnets

Page 50: An Implementation of Multiply Sectioned Bayesian Networks

50

>> setBN('benzene.dne')

The global BN is now "benzene.dne".

The CPDs are:------------------ ------------------------ ------------------------| P(A) | | P(C|A) | | P(B|A) | |----------------| |----------------------| |----------------------| | A | Values | | A C | Values | | A B | Values | |----------------| |----------------------| |----------------------| | True | 0.3000 | | True True | 0.7500 | | True True | 1.0000 | | False | 0.7000 | | True False | 0.2500 | | True False | 0.0000 | ------------------ | False True | 0.1000 | | False True | 0.3000 |

| False False | 0.9000 | | False False | 0.7000 | ------------------------ ------------------------

------------------------ ------------------------ ------------------------------| P(E|C) | | P(D|B) | | P(F|D,E) | |----------------------| |----------------------| |----------------------------| | C E | Values | | B D | Values | | E D F | Values | |----------------------| |----------------------| |----------------------------| | True True | 0.2000 | | True True | 0.2500 | | True True True | 0.1000 | | True False | 0.8000 | | True False | 0.7500 | | True True False | 0.9000 | | False True | 0.6000 | | False True | 0.3000 | | True False True | 0.3000 | | False False | 0.4000 | | False False | 0.7000 | | True False False | 0.7000 | ------------------------ ------------------------ | False True True | 0.2000 |

| False True False | 0.8000 | Description of cliques: | False False True | 0.4000 | Clique 1 has nodes A, C, B. | False False False | 0.6000 | Clique 2 has nodes C, B, E. ------------------------------Clique 3 has nodes B, E, D.Clique 4 has nodes E, D, F.

MatLab ToolMatLab Tool

• Bayes Net

– variables and states

– CPDs

– entered as simple text format or translated from a .dne Netica file

• Junction tree from Bayes Net

– entered as simple text format or translated from a .dne Netica file

• Partition of the junction tree

Inputs

Page 51: An Implementation of Multiply Sectioned Bayesian Networks

51

i + [Enter] to Initialize, v + [Enter] to view all variables (even those

containing no information),e + [Enter] to enter evidence, c + [Enter] to perform a inter-subnet

communication,p + [Enter] to go to the previous step, n + [Enter] to go to the next step,a + [Enter] to add a sensor, r + [Enter] to remove a sensor,t + [Enter] to turn true marginals view ON, m + [Enter] to turn discrepancy marking OFF,s + [Enter] to save to a file, q + [Enter] to quit.

Enter Command:

MatLab ToolMatLab Tool

• Insert evidence into given subnets and propagate their impact inside the subnet

• Initiate communication between subnets, followed by the propagation of new information

• View the marginal distributions of the different subnets at every step

• Step forward and backward• Save eye-friendly logs to a

file

Main functionality

Page 52: An Implementation of Multiply Sectioned Bayesian Networks

52

MatLab Tool: DisplayMatLab Tool: Display* Configuration 2: After evidence L(e|C) = (2,5) has been entered into subnet number 2

The TRUTH:------------------ ------------------ ------------------ ------------------ ------------------ ------------------| A | Values | | C | Values | | B | Values | | E | Values | | D | Values | | F | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.2005 | | True | 0.1434 | | True | 0.4403 | | True | 0.5426 | | True | 0.2780 | | True | 0.2901 | | False | 0.7995 | | False | 0.8566 | | False | 0.5597 | | False | 0.4574 | | False | 0.7220 | | False | 0.7099 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 1 (adjacent to subnets 2): Err(ACB) = 0.0527.~~~~ AD = 0.0704 / ~~~~ AD = 0.1072 / ~~~~ AD = 0.0493 / ------------------ ------------------ ------------------/ A / Values / / C / Values / / B / Values / | E | Values | | D | Values | | F | Values | /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| |----------------| |----------------| / True / 0.3000 / / True / 0.2950 / / True / 0.5100 / | True | ###### | | True | ###### | | True | ###### | / False / 0.7000 / / False / 0.7050 / / False / 0.4900 / | False | ###### | | False | ###### | | False | ###### | ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1, 3):------------------ ------------------ ------------------ ------------------ ------------------ ------------------| A | Values | | C | Values | | B | Values | | E | Values | | D | Values | | F | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | ###### | | True | 0.1434 | | True | 0.4403 | | True | 0.5426 | | True | 0.2780 | | True | ###### | | False | ###### | | False | 0.8566 | | False | 0.5597 | | False | 0.4574 | | False | 0.7220 | | False | ###### | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 2): Err(EDF) = 0.0169.------------------ ------------------ ------------------ ~~~~ AD = 0.0429 / ~~~~ AD = 0.0025 / ~~~~ AD = 0.0048 / | A | Values | | C | Values | | B | Values | / E / Values / / D / Values / / F / Values / |----------------| |----------------| |----------------| /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ | True | ###### | | True | ###### | | True | ###### | / True / 0.4820 / / True / 0.2745 / / True / 0.2969 / | False | ###### | | False | ###### | | False | ###### | / False / 0.5180 / / False / 0.7255 / / False / 0.7031 / ------------------ ------------------ ------------------ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

Enter a command (enter h + [Enter] for help):

Indicates step number and last action that was taken

Shows the marginal distributions that would have been obtained by infering on the entire Bayes Net

Shows the marginal distributions of the variables represented in each subnet

Prompts for new action

Page 53: An Implementation of Multiply Sectioned Bayesian Networks

53

MatLab Tool: DisplayMatLab Tool: Display* Configuration 2: After evidence L(e|C) = (2,5) has been entered into subnet number 2

The TRUTH:------------------ ------------------ ------------------ ------------------ ------------------ ------------------| A | Values | | C | Values | | B | Values | | E | Values | | D | Values | | F | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.2005 | | True | 0.1434 | | True | 0.4403 | | True | 0.5426 | | True | 0.2780 | | True | 0.2901 | | False | 0.7995 | | False | 0.8566 | | False | 0.5597 | | False | 0.4574 | | False | 0.7220 | | False | 0.7099 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 1 (adjacent to subnets 2): Err(ACB) = 0.0527.~~~~ AD = 0.0704 / ~~~~ AD = 0.1072 / ~~~~ AD = 0.0493 / ------------------ ------------------ ------------------/ A / Values / / C / Values / / B / Values / | E | Values | | D | Values | | F | Values | /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| |----------------| |----------------| / True / 0.3000 / / True / 0.2950 / / True / 0.5100 / | True | ###### | | True | ###### | | True | ###### | / False / 0.7000 / / False / 0.7050 / / False / 0.4900 / | False | ###### | | False | ###### | | False | ###### | ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1, 3):------------------ ------------------ ------------------ ------------------ ------------------ ------------------| A | Values | | C | Values | | B | Values | | E | Values | | D | Values | | F | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | ###### | | True | 0.1434 | | True | 0.4403 | | True | 0.5426 | | True | 0.2780 | | True | ###### | | False | ###### | | False | 0.8566 | | False | 0.5597 | | False | 0.4574 | | False | 0.7220 | | False | ###### | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 2): Err(EDF) = 0.0169.------------------ ------------------ ------------------ ~~~~ AD = 0.0429 / ~~~~ AD = 0.0025 / ~~~~ AD = 0.0048 / | A | Values | | C | Values | | B | Values | / E / Values / / D / Values / / F / Values / |----------------| |----------------| |----------------| /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ | True | ###### | | True | ###### | | True | ###### | / True / 0.4820 / / True / 0.2745 / / True / 0.2969 / | False | ###### | | False | ###### | | False | ###### | / False / 0.5180 / / False / 0.7255 / / False / 0.7031 / ------------------ ------------------ ------------------ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

Enter a command (enter h + [Enter] for help): Indicates that subnet does not represent this variable

Indicates that marginal variable distribution matches “true marginal”

Indicates that marginal variable distribution differs from “true marginal”

Shows average discrapancy between subnet marginals and “true marginals”

Shows average discrepancy between subnet and true joint distributions of the variables of the subnet

Page 54: An Implementation of Multiply Sectioned Bayesian Networks

54

MatLab ToolMatLab Tool

• Bayes Net

– variables and states

– CPDs

– entered as simple text format or translated from a .dne Netica file

• Junction tree from Bayes Net

– entered as simple text format or translated from a .dne Netica file

• Partition of the junction tree

>> setBN('benzene.dne')

The global BN is now "benzene.dne".

The CPDs are: ------------------ ------------------------ ------------------------ | P(A) | | P(C|A) | | P(B|A) | |----------------| |----------------------| |----------------------| | A | Values | | A C | Values | | A B | Values | |----------------| |----------------------| |----------------------| | True | 0.3000 | | True True | 0.7500 | | True True | 1.0000 | | False | 0.7000 | | True False | 0.2500 | | True False | 0.0000 | ------------------ | False True | 0.1000 | | False True | 0.3000 | | False False | 0.9000 | | False False | 0.7000 | ------------------------ ------------------------

------------------------ ------------------------ ------------------------------ | P(E|C) | | P(D|B) | | P(F|D,E) | |----------------------| |----------------------| |----------------------------| | C E | Values | | B D | Values | | E D F | Values | |----------------------| |----------------------| |----------------------------| | True True | 0.2000 | | True True | 0.2500 | | True True True | 0.1000 | | True False | 0.8000 | | True False | 0.7500 | | True True False | 0.9000 | | False True | 0.6000 | | False True | 0.3000 | | True False True | 0.3000 | | False False | 0.4000 | | False False | 0.7000 | | True False False | 0.7000 | ------------------------ ------------------------ | False True True | 0.2000 | | False True False | 0.8000 | Description of cliques: | False False True | 0.4000 | Clique 1 has nodes A, C, B. | False False False | 0.6000 | Clique 2 has nodes C, B, E. ------------------------------ Clique 3 has nodes B, E, D.Clique 4 has nodes E, D, F.

Inputs

Page 55: An Implementation of Multiply Sectioned Bayesian Networks

55

i + [Enter] to Initialize,

v + [Enter] to view all variables (even those containing no information),

e + [Enter] to enter evidence,

c + [Enter] to perform a inter-subnet communication,

p + [Enter] to go to the previous step,

n + [Enter] to go to the next step,a + [Enter] to add a sensor,

r + [Enter] to remove a sensor,t + [Enter] to turn true marginals view ON,

m + [Enter] to turn discrepancy marking

OFF,s + [Enter] to save to a file,

q + [Enter] to quit.

Enter Command:

MatLab ToolMatLab Tool

• Insert evidence into given subnets and propagate their impact inside the subnet

• Initiate communication between subnets, followed by the propagation of new information

• View the marginal distributions of the different subnets at every step

• Step forward and backward• Save eye-friendly logs to a

file

Main functionality

Page 56: An Implementation of Multiply Sectioned Bayesian Networks

56

MatLab Tool: DisplayMatLab Tool: Display * Configuration 2: After evidence L(e|C) = (2,5) has been entered into subnet number 2

The TRUTH:

------------------ ------------------ ------------------ ------------------ ------------------ ------------------

| A | Values | | C | Values | | B | Values | | E | Values | | D | Values | | F | Values |

|----------------| |----------------| |----------------| |----------------| |----------------| |----------------|

| True | 0.2005 | | True | 0.1434 | | True | 0.4403 | | True | 0.5426 | | True | 0.2780 | | True | 0.2901 |

| False | 0.7995 | | False | 0.8566 | | False | 0.5597 | | False | 0.4574 | | False | 0.7220 | | False | 0.7099 |

------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 1 (adjacent to subnets 2): Err(ACB) = 0.0527.

~~~~ AD = 0.0704 / ~~~~ AD = 0.1072 / ~~~~ AD = 0.0493 / ------------------ ------------------ ------------------

/ A / Values / / C / Values / / B / Values / | E | Values | | D | Values | | F | Values |

/~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| |----------------| |----------------|

/ True / 0.3000 / / True / 0.2950 / / True / 0.5100 / | True | ###### | | True | ###### | | True | ###### |

/ False / 0.7000 / / False / 0.7050 / / False / 0.4900 / | False | ###### | | False | ###### | | False | ###### |

~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1, 3):

------------------ ------------------ ------------------ ------------------ ------------------ ------------------

| A | Values | | C | Values | | B | Values | | E | Values | | D | Values | | F | Values |

|----------------| |----------------| |----------------| |----------------| |----------------| |----------------|

| True | ###### | | True | 0.1434 | | True | 0.4403 | | True | 0.5426 | | True | 0.2780 | | True | ###### |

| False | ###### | | False | 0.8566 | | False | 0.5597 | | False | 0.4574 | | False | 0.7220 | | False | ###### |

------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 2): Err(EDF) = 0.0169.

------------------ ------------------ ------------------ ~~~~ AD = 0.0429 / ~~~~ AD = 0.0025 / ~~~~ AD = 0.0048 /

| A | Values | | C | Values | | B | Values | / E / Values / / D / Values / / F / Values /

|----------------| |----------------| |----------------| /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/

| True | ###### | | True | ###### | | True | ###### | / True / 0.4820 / / True / 0.2745 / / True / 0.2969 /

| False | ###### | | False | ###### | | False | ###### | / False / 0.5180 / / False / 0.7255 / / False / 0.7031 /

------------------ ------------------ ------------------ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

Enter a command (enter h + [Enter] for help):

Indicates step number and last action that was taken

Shows the marginal distributions that would have been obtained by infering on the entire Bayes Net

Shows the marginal distributions of the variables represented in each subnet

Prompts for new action

Page 57: An Implementation of Multiply Sectioned Bayesian Networks

57

MatLab Tool: DisplayMatLab Tool: Display * Configuration 2: After evidence L(e|C) = (2,5) has been entered into subnet number 2

The TRUTH:

------------------ ------------------ ------------------ ------------------ ------------------ ------------------

| A | Values | | C | Values | | B | Values | | E | Values | | D | Values | | F | Values |

|----------------| |----------------| |----------------| |----------------| |----------------| |----------------|

| True | 0.2005 | | True | 0.1434 | | True | 0.4403 | | True | 0.5426 | | True | 0.2780 | | True | 0.2901 |

| False | 0.7995 | | False | 0.8566 | | False | 0.5597 | | False | 0.4574 | | False | 0.7220 | | False | 0.7099 |

------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 1 (adjacent to subnets 2): Err(ACB) = 0.0527.

~~~~ AD = 0.0704 / ~~~~ AD = 0.1072 / ~~~~ AD = 0.0493 / ------------------ ------------------ ------------------

/ A / Values / / C / Values / / B / Values / | E | Values | | D | Values | | F | Values |

/~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| |----------------| |----------------|

/ True / 0.3000 / / True / 0.2950 / / True / 0.5100 / | True | ###### | | True | ###### | | True | ###### |

/ False / 0.7000 / / False / 0.7050 / / False / 0.4900 / | False | ###### | | False | ###### | | False | ###### |

~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1, 3):

------------------ ------------------ ------------------ ------------------ ------------------ ------------------

| A | Values | | C | Values | | B | Values | | E | Values | | D | Values | | F | Values |

|----------------| |----------------| |----------------| |----------------| |----------------| |----------------|

| True | ###### | | True | 0.1434 | | True | 0.4403 | | True | 0.5426 | | True | 0.2780 | | True | ###### |

| False | ###### | | False | 0.8566 | | False | 0.5597 | | False | 0.4574 | | False | 0.7220 | | False | ###### |

------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 2): Err(EDF) = 0.0169.

------------------ ------------------ ------------------ ~~~~ AD = 0.0429 / ~~~~ AD = 0.0025 / ~~~~ AD = 0.0048 /

| A | Values | | C | Values | | B | Values | / E / Values / / D / Values / / F / Values /

|----------------| |----------------| |----------------| /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/

| True | ###### | | True | ###### | | True | ###### | / True / 0.4820 / / True / 0.2745 / / True / 0.2969 /

| False | ###### | | False | ###### | | False | ###### | / False / 0.5180 / / False / 0.7255 / / False / 0.7031 /

------------------ ------------------ ------------------ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

Enter a command (enter h + [Enter] for help):

Indicates that subnet does not represent this variable

Indicates that marginal variable distribution differs from “true marginal”

Shows average discrapancy between subnet marginals and “true marginals”

Shows average discrepancy between subnet and true joint distributions of the variables of the subnet

Indicates that marginal variable distribution matches “true marginal”

Page 58: An Implementation of Multiply Sectioned Bayesian Networks

58

MatLab Tool: ControlMatLab Tool: ControlEnter a command (enter h + [Enter] for help): hType:

i + [Enter] to Initialize, v + [Enter] to view all variables (even those containing no information),e + [Enter] to enter evidence, c + [Enter] to perform a inter-subnet communication,

p + [Enter] to go to the previous step, n + [Enter] to go to the next step,

a + [Enter] to add a sensor, r + [Enter] to remove a sensor,t + [Enter] to turn true marginals view OFF, m + [Enter] to turn discrepancy marking OFF,s + [Enter] to save to a file, q + [Enter] to quit.

So, what do you want to do?

Page 59: An Implementation of Multiply Sectioned Bayesian Networks

59

MatLab Tool: Initial stateMatLab Tool: Initial state * Configuration 1: Initial configuration:

SUBNET 1 (adjacent to subnets 2, 3, 4, 5): ------------------ ------------------ ------------------ ------------------ ------------------ ------------------ | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | | RF5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.0069 | | True | 0.8064 | | True | 0.6287 | | True | 0.5420 | | True | 0.2257 | | True | 0.4738 | | False | 0.9931 | | False | 0.1936 | | False | 0.3713 | | False | 0.4580 | | False | 0.7743 | | False | 0.5262 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2257 | | True | 0.8332 | | True | 0.4738 | | False | 0.7743 | | False | 0.1668 | | False | 0.5262 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ ------------------ | CF3 | Values | | RF5 | Values | | RF4 | Values | | RM6 | Values | | RM5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.5420 | | True | 0.4738 | | True | 0.4616 | | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | | False | 0.5262 | | False | 0.5384 | | False | 0.5000 | | False | 0.7798 | ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8064 | | True | 0.6287 | | True | 0.2472 | | True | 0.5120 | | False | 0.1936 | | False | 0.3713 | | False | 0.7528 | | False | 0.4880 | ------------------ ------------------ ------------------ ------------------

SUBNET 5 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM3 | Values | | RM4 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8064 | | True | 0.6287 | | True | 0.4481 | | True | 0.3351 | | False | 0.1936 | | False | 0.3713 | | False | 0.5519 | | False | 0.6649 | ------------------ ------------------ ------------------ ------------------

Page 60: An Implementation of Multiply Sectioned Bayesian Networks

60

MatLab Tool: Initial stateMatLab Tool: Initial state * Configuration 1: Initial configuration:

SUBNET 1 (adjacent to subnets 2, 3, 4, 5): ------------------ ------------------ ------------------ ------------------ ------------------ ------------------ | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | | RF5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.0069 | | True | 0.8064 | | True | 0.6287 | | True | 0.5420 | | True | 0.2257 | | True | 0.4738 | | False | 0.9931 | | False | 0.1936 | | False | 0.3713 | | False | 0.4580 | | False | 0.7743 | | False | 0.5262 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2257 | | True | 0.8332 | | True | 0.4738 | | False | 0.7743 | | False | 0.1668 | | False | 0.5262 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ ------------------ | CF3 | Values | | RF5 | Values | | RF4 | Values | | RM6 | Values | | RM5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.5420 | | True | 0.4738 | | True | 0.4616 | | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | | False | 0.5262 | | False | 0.5384 | | False | 0.5000 | | False | 0.7798 | ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8064 | | True | 0.6287 | | True | 0.2472 | | True | 0.5120 | | False | 0.1936 | | False | 0.3713 | | False | 0.7528 | | False | 0.4880 | ------------------ ------------------ ------------------ ------------------

SUBNET 5 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM3 | Values | | RM4 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8064 | | True | 0.6287 | | True | 0.4481 | | True | 0.3351 | | False | 0.1936 | | False | 0.3713 | | False | 0.5519 | | False | 0.6649 | ------------------ ------------------ ------------------ ------------------

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

RF1

RM3RM4

RF2

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

2345

Page 61: An Implementation of Multiply Sectioned Bayesian Networks

61

MatLab Tool: Remove a sensorMatLab Tool: Remove a sensor * Configuration 2: The radar-subnet number 5 was removed.

SUBNET 1 (adjacent to subnets 2, 3, 4): ------------------ ------------------ ------------------ ------------------ ------------------ ------------------ | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | | RF5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.0069 | | True | 0.8064 | | True | 0.6287 | | True | 0.5420 | | True | 0.2257 | | True | 0.4738 | | False | 0.9931 | | False | 0.1936 | | False | 0.3713 | | False | 0.4580 | | False | 0.7743 | | False | 0.5262 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2257 | | True | 0.8332 | | True | 0.4738 | | False | 0.7743 | | False | 0.1668 | | False | 0.5262 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ ------------------ | CF3 | Values | | RF5 | Values | | RF4 | Values | | RM6 | Values | | RM5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.5420 | | True | 0.4738 | | True | 0.4616 | | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | | False | 0.5262 | | False | 0.5384 | | False | 0.5000 | | False | 0.7798 | ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8064 | | True | 0.6287 | | True | 0.2472 | | True | 0.5120 | | False | 0.1936 | | False | 0.3713 | | False | 0.7528 | | False | 0.4880 | ------------------ ------------------ ------------------ ------------------

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

234

Page 62: An Implementation of Multiply Sectioned Bayesian Networks

62

MatLab Tool: Receive evidenceMatLab Tool: Receive evidenceEnter a command (enter h + [Enter] for help): e

In which subnet (number) do you wish to insert evidence? 2

The variables handled by subnet number 2 of this cell are: 5: CM7 6: RM8 7: RF5 Which variable NUMBER does the evidence pertain to? 6

Enter likelihood for state "True": 0

Enter likelihood for state "False": 1

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

234

Subnet 2 receives evidence on RM8.

Page 63: An Implementation of Multiply Sectioned Bayesian Networks

63

MatLab Tool: Receive evidenceMatLab Tool: Receive evidence * Configuration 3: After evidence L(e|RM8) = (0,1) has been entered into subnet number 2

SUBNET 1 (adjacent to subnets 2, 3, 4): Err(CTCF3RF5) = 0.0368. Err(CTCM7RF5) = 0.0395. ------------------ ------------------ ------------------ ------------------ ------------------ ~~~~ AD = 0.0784 / | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | / RF5 / Values / |----------------| |----------------| |----------------| |----------------| |----------------| /~~~~~~~~~~~~~~~~/ | True | 0.0069 | | True | 0.8064 | | True | 0.6287 | | True | 0.5420 | | True | 0.2257 | / True / 0.4738 / | False | 0.9931 | | False | 0.1936 | | False | 0.3713 | | False | 0.4580 | | False | 0.7743 | / False / 0.5262 / ------------------ ------------------ ------------------ ------------------ ------------------ ~~~~~~~~~~~~~~~~~~

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2257 | | True | 0.0000 | | True | 0.3629 | | False | 0.7743 | | False | 1.0000 | | False | 0.6371 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): Err(CF3RM5RF5RF4) = 0.0169. Err(RF5RM6) = 0.0555. ------------------ ~~~~ AD = 0.0784 / ~~~~ AD = 0.0119 / ------------------ ------------------ | CF3 | Values | / RF5 / Values / / RF4 / Values / | RM6 | Values | | RM5 | Values | |----------------| /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| |----------------| | True | 0.5420 | / True / 0.4738 / / True / 0.4616 / | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | / False / 0.5262 / / False / 0.5384 / | False | 0.5000 | | False | 0.7798 | ------------------ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8064 | | True | 0.6287 | | True | 0.2472 | | True | 0.5120 | | False | 0.1936 | | False | 0.3713 | | False | 0.7528 | | False | 0.4880 | ------------------ ------------------ ------------------ ------------------

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

234

Page 64: An Implementation of Multiply Sectioned Bayesian Networks

64

MatLab Tool: Receive evidenceMatLab Tool: Receive evidenceEnter a command (enter h + [Enter] for help): e

In which subnet (number) do you wish to insert evidence? 4

The variables handled by subnet number 4 of this cell are: 2: RF1 3: RF2 11: RM1 12: RM2 Which variable NUMBER does the evidence pertain to? 12

Enter likelihood for state "True": 3

Enter likelihood for state "False": 5

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

234

Subnet 4 receives evidence on RM2.

Page 65: An Implementation of Multiply Sectioned Bayesian Networks

65

MatLab Tool: Receive evidenceMatLab Tool: Receive evidence * Configuration 4: After evidence L(e|RM2) = (3,5) has been entered into subnet number 4

SUBNET 1 (adjacent to subnets 2, 3, 4): Err(CTRF1RF2) = 0.0059. Err(CTCF3RF5) = 0.0368. Err(CTCM7RF5) = 0.0395. ------------------ ~~~~ AD = 0.0012 / ~~~~ AD = 0.0063 / ------------------ ------------------ ~~~~ AD = 0.0784 / | CT | Values | / RF1 / Values / / RF2 / Values / | CF3 | Values | | CM7 | Values | / RF5 / Values / |----------------| /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| |----------------| /~~~~~~~~~~~~~~~~/ | True | 0.0069 | / True / 0.8064 / / True / 0.6287 / | True | 0.5420 | | True | 0.2257 | / True / 0.4738 / | False | 0.9931 | / False / 0.1936 / / False / 0.3713 / | False | 0.4580 | | False | 0.7743 | / False / 0.5262 / ------------------ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ------------------ ~~~~~~~~~~~~~~~~~~

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2257 | | True | 0.0000 | | True | 0.3629 | | False | 0.7743 | | False | 1.0000 | | False | 0.6371 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): Err(CF3RM5RF5RF4) = 0.0169. Err(RF5RM6) = 0.0555. ------------------ ~~~~ AD = 0.0784 / ~~~~ AD = 0.0119 / ------------------ ------------------ | CF3 | Values | / RF5 / Values / / RF4 / Values / | RM6 | Values | | RM5 | Values | |----------------| /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| |----------------| | True | 0.5420 | / True / 0.4738 / / True / 0.4616 / | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | / False / 0.5262 / / False / 0.5384 / | False | 0.5000 | | False | 0.7798 | ------------------ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8081 | | True | 0.6376 | | True | 0.2633 | | True | 0.3863 | | False | 0.1919 | | False | 0.3624 | | False | 0.7367 | | False | 0.6137 | ------------------ ------------------ ------------------ ------------------

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

234

Page 66: An Implementation of Multiply Sectioned Bayesian Networks

66

* Configuration 5: After subnet 4 sent information to subnet 1

SUBNET 1 (adjacent to subnets 2, 3, 4): Err(CTCF3RF5) = 0.0368. Err(CTCM7RF5) = 0.0395. ------------------ ------------------ ------------------ ------------------ ------------------ ~~~~ AD = 0.0784 / | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | / RF5 / Values / |----------------| |----------------| |----------------| |----------------| |----------------| /~~~~~~~~~~~~~~~~/ | True | 0.0068 | | True | 0.8081 | | True | 0.6376 | | True | 0.5420 | | True | 0.2257 | / True / 0.4738 / | False | 0.9932 | | False | 0.1919 | | False | 0.3624 | | False | 0.4580 | | False | 0.7743 | / False / 0.5262 / ------------------ ------------------ ------------------ ------------------ ------------------ ~~~~~~~~~~~~~~~~~~

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2257 | | True | 0.0000 | | True | 0.3629 | | False | 0.7743 | | False | 1.0000 | | False | 0.6371 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): Err(CF3RM5RF5RF4) = 0.0169. Err(RF5RM6) = 0.0555. ------------------ ~~~~ AD = 0.0784 / ~~~~ AD = 0.0119 / ------------------ ------------------ | CF3 | Values | / RF5 / Values / / RF4 / Values / | RM6 | Values | | RM5 | Values | |----------------| /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| |----------------| | True | 0.5420 | / True / 0.4738 / / True / 0.4616 / | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | / False / 0.5262 / / False / 0.5384 / | False | 0.5000 | | False | 0.7798 | ------------------ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8081 | | True | 0.6376 | | True | 0.2633 | | True | 0.3863 | | False | 0.1919 | | False | 0.3624 | | False | 0.7367 | | False | 0.6137 | ------------------ ------------------ ------------------ ------------------

MatLab Tool: Subnet communicationMatLab Tool: Subnet communication

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

234Enter a command (enter h + [Enter] for help): c Send information FROM subnet number: 4 Subnet 4 can communicate with the following subnets: 1 Enter the subnet number that subnet 4 should send information

TO: 1

Page 67: An Implementation of Multiply Sectioned Bayesian Networks

67

* Configuration 6: After subnet 2 sent information to subnet 1

SUBNET 1 (adjacent to subnets 2, 3, 4): ------------------ ------------------ ------------------ ------------------ ------------------ ------------------ | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | | RF5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.0068 | | True | 0.8081 | | True | 0.6376 | | True | 0.5420 | | True | 0.2257 | | True | 0.3628 | | False | 0.9932 | | False | 0.1919 | | False | 0.3624 | | False | 0.4580 | | False | 0.7743 | | False | 0.6372 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2257 | | True | 0.0000 | | True | 0.3629 | | False | 0.7743 | | False | 1.0000 | | False | 0.6371 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): Err(CF3RM5RF5RF4) = 0.0169. Err(RF5RM6) = 0.0555. ------------------ ~~~~ AD = 0.0784 / ~~~~ AD = 0.0119 / ------------------ ------------------ | CF3 | Values | / RF5 / Values / / RF4 / Values / | RM6 | Values | | RM5 | Values | |----------------| /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| |----------------| | True | 0.5420 | / True / 0.4738 / / True / 0.4616 / | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | / False / 0.5262 / / False / 0.5384 / | False | 0.5000 | | False | 0.7798 | ------------------ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8081 | | True | 0.6376 | | True | 0.2633 | | True | 0.3863 | | False | 0.1919 | | False | 0.3624 | | False | 0.7367 | | False | 0.6137 | ------------------ ------------------ ------------------ ------------------

MatLab Tool: Subnet communicationMatLab Tool: Subnet communication

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

234Enter a command (enter h + [Enter] for help): c Send information FROM subnet number: 2 Subnet 4 can communicate with the following subnets: 1 Enter the subnet number that subnet 4 should send information

TO: 1

Page 68: An Implementation of Multiply Sectioned Bayesian Networks

68

* Configuration 7: After subnet 1 sent information to subnet 3

SUBNET 1 (adjacent to subnets 2, 3, 4): ------------------ ------------------ ------------------ ------------------ ------------------ ------------------ | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | | RF5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.0068 | | True | 0.8081 | | True | 0.6376 | | True | 0.5420 | | True | 0.2257 | | True | 0.3628 | | False | 0.9932 | | False | 0.1919 | | False | 0.3624 | | False | 0.4580 | | False | 0.7743 | | False | 0.6372 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2257 | | True | 0.0000 | | True | 0.3629 | | False | 0.7743 | | False | 1.0000 | | False | 0.6371 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ ------------------ | CF3 | Values | | RF5 | Values | | RF4 | Values | | RM6 | Values | | RM5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.5420 | | True | 0.3628 | | True | 0.4784 | | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | | False | 0.6372 | | False | 0.5216 | | False | 0.5000 | | False | 0.7798 | ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8081 | | True | 0.6376 | | True | 0.2633 | | True | 0.3863 | | False | 0.1919 | | False | 0.3624 | | False | 0.7367 | | False | 0.6137 | ------------------ ------------------ ------------------ ------------------

MatLab Tool: Subnet communicationMatLab Tool: Subnet communication

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

234Enter a command (enter h + [Enter] for help): c Send information FROM subnet number: 1 Subnet 4 can communicate with the following subnets: 2, 3, 4 Enter the subnet number that subnet 4 should send information TO:

3

Page 69: An Implementation of Multiply Sectioned Bayesian Networks

69

MatLab Tool: Add a subnetMatLab Tool: Add a subnet * Configuration 8: The radar-subnet number 5 was added.

SUBNET 1 (adjacent to subnets 2, 3, 4, 5): Err(CTRF1RF2) = 0.0227. Err(CTCF3RF5) = 0.0004. Err(CTCM7RF5) = 0.0004. ~~~~ AD = 0.0011 / ~~~~ AD = 0.0464 / ~~~~ AD = 0.0087 / ------------------ ~~~~ AD = 0.0002 / ------------------ / CT / Values / / RF1 / Values / / RF2 / Values / | CF3 | Values | / CM7 / Values / | RF5 | Values | /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ |----------------| /~~~~~~~~~~~~~~~~/ |----------------| / True / 0.0068 / / True / 0.8081 / / True / 0.6376 / | True | 0.5420 | / True / 0.2257 / | True | 0.3628 | / False / 0.9932 / / False / 0.1919 / / False / 0.3624 / | False | 0.4580 | / False / 0.7743 / | False | 0.6372 | ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ------------------ ~~~~~~~~~~~~~~~~~~ ------------------

SUBNET 2 (adjacent to subnets 1): Err(CM7RM8RF5) = 0.0001. ~~~~ AD = 0.0002 / ------------------ ------------------ / CM7 / Values / | RM8 | Values | | RF5 | Values | /~~~~~~~~~~~~~~~~/ |----------------| |----------------| / True / 0.2257 / | True | 0.0000 | | True | 0.3629 | / False / 0.7743 / | False | 1.0000 | | False | 0.6371 | ~~~~~~~~~~~~~~~~~~ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ ------------------ | CF3 | Values | | RF5 | Values | | RF4 | Values | | RM6 | Values | | RM5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.5420 | | True | 0.3628 | | True | 0.4784 | | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | | False | 0.6372 | | False | 0.5216 | | False | 0.5000 | | False | 0.7798 | ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): Err(RF1RF2RM1RM2) = 0.0099. ~~~~ AD = 0.0464 / ~~~~ AD = 0.0087 / ~~~~ AD = 0.0009 / ~~~~ AD = 0.0011 / / RF1 / Values / / RF2 / Values / / RM1 / Values / / RM2 / Values / /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ / True / 0.8081 / / True / 0.6376 / / True / 0.2633 / / True / 0.3863 / / False / 0.1919 / / False / 0.3624 / / False / 0.7367 / / False / 0.6137 / ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

SUBNET 5 (adjacent to subnets 1): Err(RF1RF2RM3RM4) = 0.2284. ~~~~ AD = 0.1826 / ~~~~ AD = 0.1091 / ~~~~ AD = 0.0122 / ~~~~ AD = 0.0123 / / RF1 / Values / / RF2 / Values / / RM3 / Values / / RM4 / Values / /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ / True / 0.6154 / / True / 0.4710 / / True / 0.1905 / / True / 0.5625 / / False / 0.3846 / / False / 0.5290 / / False / 0.8095 / / False / 0.4375 / ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

RF1

RM3RM4

RF2

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

2345

Enter a command (enter h + [Enter] for help): a

Page 70: An Implementation of Multiply Sectioned Bayesian Networks

70

MatLab Tool: Subnet communicationMatLab Tool: Subnet communication * Configuration 9: After subnet 5 sent information to subnet 1

SUBNET 1 (adjacent to subnets 2, 3, 4, 5): ------------------ ------------------ ------------------ ------------------ ------------------ ------------------ | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | | RF5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.0052 | | True | 0.8736 | | True | 0.6253 | | True | 0.5419 | | True | 0.2254 | | True | 0.3628 | | False | 0.9948 | | False | 0.1264 | | False | 0.3747 | | False | 0.4581 | | False | 0.7746 | | False | 0.6372 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1): Err(CM7RM8RF5) = 0.0001. ~~~~ AD = 0.0002 / ------------------ ------------------ / CM7 / Values / | RM8 | Values | | RF5 | Values | /~~~~~~~~~~~~~~~~/ |----------------| |----------------| / True / 0.2257 / | True | 0.0000 | | True | 0.3629 | / False / 0.7743 / | False | 1.0000 | | False | 0.6371 | ~~~~~~~~~~~~~~~~~~ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ ------------------ | CF3 | Values | | RF5 | Values | | RF4 | Values | | RM6 | Values | | RM5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.5420 | | True | 0.3628 | | True | 0.4784 | | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | | False | 0.6372 | | False | 0.5216 | | False | 0.5000 | | False | 0.7798 | ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): Err(RF1RF2RM1RM2) = 0.0099. ~~~~ AD = 0.0464 / ~~~~ AD = 0.0087 / ~~~~ AD = 0.0009 / ~~~~ AD = 0.0011 / / RF1 / Values / / RF2 / Values / / RM1 / Values / / RM2 / Values / /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ / True / 0.8081 / / True / 0.6376 / / True / 0.2633 / / True / 0.3863 / / False / 0.1919 / / False / 0.3624 / / False / 0.7367 / / False / 0.6137 / ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

SUBNET 5 (adjacent to subnets 1): Err(RF1RF2RM3RM4) = 0.2284. ~~~~ AD = 0.1826 / ~~~~ AD = 0.1091 / ~~~~ AD = 0.0122 / ~~~~ AD = 0.0123 / / RF1 / Values / / RF2 / Values / / RM3 / Values / / RM4 / Values / /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ / True / 0.6154 / / True / 0.4710 / / True / 0.1905 / / True / 0.5625 / / False / 0.3846 / / False / 0.5290 / / False / 0.8095 / / False / 0.4375 / ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

RF1

RM3RM4

RF2

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

2345

Page 71: An Implementation of Multiply Sectioned Bayesian Networks

71

MatLab Tool: Subnet communicationMatLab Tool: Subnet communication * Configuration 10: After subnet 1 sent information to subnet 2

SUBNET 1 (adjacent to subnets 2, 3, 4, 5): ------------------ ------------------ ------------------ ------------------ ------------------ ------------------ | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | | RF5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.0052 | | True | 0.8736 | | True | 0.6253 | | True | 0.5419 | | True | 0.2254 | | True | 0.3628 | | False | 0.9948 | | False | 0.1264 | | False | 0.3747 | | False | 0.4581 | | False | 0.7746 | | False | 0.6372 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2254 | | True | 0.0000 | | True | 0.3628 | | False | 0.7746 | | False | 1.0000 | | False | 0.6372 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ ------------------ | CF3 | Values | | RF5 | Values | | RF4 | Values | | RM6 | Values | | RM5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.5420 | | True | 0.3628 | | True | 0.4784 | | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | | False | 0.6372 | | False | 0.5216 | | False | 0.5000 | | False | 0.7798 | ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): Err(RF1RF2RM1RM2) = 0.0099. ~~~~ AD = 0.0464 / ~~~~ AD = 0.0087 / ~~~~ AD = 0.0009 / ~~~~ AD = 0.0011 / / RF1 / Values / / RF2 / Values / / RM1 / Values / / RM2 / Values / /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ / True / 0.8081 / / True / 0.6376 / / True / 0.2633 / / True / 0.3863 / / False / 0.1919 / / False / 0.3624 / / False / 0.7367 / / False / 0.6137 / ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

SUBNET 5 (adjacent to subnets 1): Err(RF1RF2RM3RM4) = 0.2284. ~~~~ AD = 0.1826 / ~~~~ AD = 0.1091 / ~~~~ AD = 0.0122 / ~~~~ AD = 0.0123 / / RF1 / Values / / RF2 / Values / / RM3 / Values / / RM4 / Values / /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ / True / 0.6154 / / True / 0.4710 / / True / 0.1905 / / True / 0.5625 / / False / 0.3846 / / False / 0.5290 / / False / 0.8095 / / False / 0.4375 / ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

RF1

RM3RM4

RF2

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

2345

Page 72: An Implementation of Multiply Sectioned Bayesian Networks

72

MatLab Tool: Subnet communicationMatLab Tool: Subnet communication * Configuration 11: After subnet 1 sent information to subnet 4

SUBNET 1 (adjacent to subnets 2, 3, 4, 5): ------------------ ------------------ ------------------ ------------------ ------------------ ------------------ | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | | RF5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.0052 | | True | 0.8736 | | True | 0.6253 | | True | 0.5419 | | True | 0.2254 | | True | 0.3628 | | False | 0.9948 | | False | 0.1264 | | False | 0.3747 | | False | 0.4581 | | False | 0.7746 | | False | 0.6372 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2254 | | True | 0.0000 | | True | 0.3628 | | False | 0.7746 | | False | 1.0000 | | False | 0.6372 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ ------------------ | CF3 | Values | | RF5 | Values | | RF4 | Values | | RM6 | Values | | RM5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.5420 | | True | 0.3628 | | True | 0.4784 | | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | | False | 0.6372 | | False | 0.5216 | | False | 0.5000 | | False | 0.7798 | ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8736 | | True | 0.6253 | | True | 0.2645 | | True | 0.3848 | | False | 0.1264 | | False | 0.3747 | | False | 0.7355 | | False | 0.6152 | ------------------ ------------------ ------------------ ------------------

SUBNET 5 (adjacent to subnets 1): Err(RF1RF2RM3RM4) = 0.2284. ~~~~ AD = 0.1826 / ~~~~ AD = 0.1091 / ~~~~ AD = 0.0122 / ~~~~ AD = 0.0123 / / RF1 / Values / / RF2 / Values / / RM3 / Values / / RM4 / Values / /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ /~~~~~~~~~~~~~~~~/ / True / 0.6154 / / True / 0.4710 / / True / 0.1905 / / True / 0.5625 / / False / 0.3846 / / False / 0.5290 / / False / 0.8095 / / False / 0.4375 / ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

RF1

RM3RM4

RF2

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

2345

Page 73: An Implementation of Multiply Sectioned Bayesian Networks

73

MatLab Tool: Subnet communicationMatLab Tool: Subnet communication * Configuration 12: After subnet 1 sent information to subnet 5

SUBNET 1 (adjacent to subnets 2, 3, 4, 5): ------------------ ------------------ ------------------ ------------------ ------------------ ------------------ | CT | Values | | RF1 | Values | | RF2 | Values | | CF3 | Values | | CM7 | Values | | RF5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.0052 | | True | 0.8736 | | True | 0.6253 | | True | 0.5419 | | True | 0.2254 | | True | 0.3628 | | False | 0.9948 | | False | 0.1264 | | False | 0.3747 | | False | 0.4581 | | False | 0.7746 | | False | 0.6372 | ------------------ ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 2 (adjacent to subnets 1): ------------------ ------------------ ------------------ | CM7 | Values | | RM8 | Values | | RF5 | Values | |----------------| |----------------| |----------------| | True | 0.2254 | | True | 0.0000 | | True | 0.3628 | | False | 0.7746 | | False | 1.0000 | | False | 0.6372 | ------------------ ------------------ ------------------

SUBNET 3 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ ------------------ | CF3 | Values | | RF5 | Values | | RF4 | Values | | RM6 | Values | | RM5 | Values | |----------------| |----------------| |----------------| |----------------| |----------------| | True | 0.5420 | | True | 0.3628 | | True | 0.4784 | | True | 0.5000 | | True | 0.2202 | | False | 0.4580 | | False | 0.6372 | | False | 0.5216 | | False | 0.5000 | | False | 0.7798 | ------------------ ------------------ ------------------ ------------------ ------------------

SUBNET 4 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM1 | Values | | RM2 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8736 | | True | 0.6253 | | True | 0.2645 | | True | 0.3848 | | False | 0.1264 | | False | 0.3747 | | False | 0.7355 | | False | 0.6152 | ------------------ ------------------ ------------------ ------------------

SUBNET 5 (adjacent to subnets 1): ------------------ ------------------ ------------------ ------------------ | RF1 | Values | | RF2 | Values | | RM3 | Values | | RM4 | Values | |----------------| |----------------| |----------------| |----------------| | True | 0.8736 | | True | 0.6253 | | True | 0.1733 | | True | 0.5451 | | False | 0.1264 | | False | 0.3747 | | False | 0.8267 | | False | 0.4549 | ------------------ ------------------ ------------------ ------------------

CTCF3

CT

RF5

CTRF1

RF2RF1

RF5RF2

RM1RM2

RM6

RM8

RF1

RM3RM4

RF2

CM7

CM7RF5CF3

RF5RF4RM5

RF4

1

2345