[Next] [Up] [Previous]
Next: Generating Query DAGs Up: Query DAGs Previous: Implementing a Q-DAG Evaluator

The Availability of Evidence

The construction of a Q-DAG requires the identification of query and evidence variables. This may give an incorrect impression that we must know up front which variables are observed and which are not. This could be problematic in (1) applications where one may lose a sensor reading, thus changing the status of a variable from being observed to being unobserved; and (2) applications where some variable may be expensive to observe, leading to an on-line decision on whether to observe it or not (using some value-of-information computation).

 figure178
Figure: A belief network and its corresponding Q-DAG in which variable tex2html_wrap1015 is declared to be both query and evidence.  

Both of these situations can be dealt with in a Q-DAG framework. First, as we mentioned earlier, Q-DAGs allow us to handle missing evidence through the use of the tex2html_wrap1065 notation which denotes an unknown value of a variable. Therefore, Q-DAGs can handle missing sensor readings. Second, a variable can be declared to be both query and evidence. This means that we can incorporate evidence about this variable when it is available, and also compute the probability distribution of the variable in case evidence is not available. Figure [*] depicts a Q-DAG in which variable tex2html_wrap1256 is declared to be a query variable, while variable tex2html_wrap1015 is declared to be both an evidence and a query variable (both variables have tex2html_wrap1258 and tex2html_wrap1259 as their values). In this case, we have two ESNs for variable tex2html_wrap1015 and also two query nodes (see Figure [*]). This Q-DAG can be used in two ways:

  1. To compute the probability distributions of variables tex2html_wrap1256 and tex2html_wrap1015 when no evidence is available about tex2html_wrap1015. Under this situation, the values of tex2html_wrap1264 and tex2html_wrap1265 are set to 1, and we have
    eqnarray188
  2. To compute the probability of variable tex2html_wrap1256 when evidence is available about tex2html_wrap1015. For example, suppose that we observe tex2html_wrap1015 to be tex2html_wrap1259. The value of tex2html_wrap1264 will then be set to 0 and the value of tex2html_wrap1265 will be set to 1, and we have
    eqnarray190

Utility_Of_Observing

The ability to declare a variable as both an evidence and a query variable seems to be essential in applications where (1) a decision may need to be made on whether to collect evidence about some variable tex2html_wrap1015; and (2) making the decision requires knowing the probability distribution of variable tex2html_wrap1015. For example, suppose that we are using the following formula [PearlPearl1988, Page 313,] to compute the utility of observing variable tex2html_wrap1015:
displaymath1316
where tex2html_wrap1275 is the utility for the decision maker of finding that variable tex2html_wrap1015 has value tex2html_wrap1277. Suppose that tex2html_wrap1278 and tex2html_wrap1279. We can use the Q-DAG to compute the probability distribution of tex2html_wrap1015 and use it to evaluate tex2html_wrap1281:
displaymath1317
which leads us to observe variable tex2html_wrap1015. Observing tex2html_wrap1015, we find that its value is tex2html_wrap1259. We can then accommodate this evidence into the Q-DAG and continue with our analysis.


[Next] [Up] [Previous]
Next: Generating Query DAGs Up: Query DAGs Previous: Implementing a Q-DAG Evaluator

Darwiche&Provan