next up previous
Next: Comparing VE and Up: Causal Independence and Previous: Causal Independence and

An Example

This subsection illustrates VE1 by walking through an example. Consider computing the in the deputation Bayesian network shown in Figure 2. Suppose the elimination ordering is: a, b, c, , , , and . After the first step of VE1 ,


Now the procedure enters the while-loop and it sums out the variables in one by one.

After summing out a,


where .

After summing out b,


where .

After summing out c,


where .

After summing out ,


where .

After summing out ,


where .

After summing out ,


where .

Finally, after summing out ,


where . Now the procedure enters step 3, where there is nothing to do in this example. Finally, the procedure returns , which is , the required probability.



David Poole
Fri Dec 6 15:09:32 PST 1996