DBN Inference: How to compute CPT of nodes in time-slice > 1

[adusia]

Hello Everyone,

I am trying to learn DBN and solve few examples.

In this example, random CPTs are assigned to nodes: http://www.cs.ubc.ca/~murphyk/Bayes/usage_dbn.may22.html#hmm

Code: bnet = mk_dbn(intra, inter, ns, dnodes); for i=1:4 bnet.CPD{i} = tabular_CPD(bnet, i); end

I am trying to the model the example on this page as DBN: http://bnt.googlecode.com/svn/trunk/docs/usage.html#basics

1) The initial CPT for time-slice=1 is given in the example 2) I am assuming the transitional prob of all hidden nodes to be 0.9, i.e. P(Xt/ X(t-1)) = 0.9

Now my question is how to compute the CPT of C2, S2, R2 for time-slice 2?

S2 here will depend on S1 and C2

I would really appreciate any kind of help.

Thanks in advance!

AD