** Next:** About this document ...
** Up:** Computing the Optimal Policy
** Previous:** Uniqueness of

##

Example:

* Using the same example we calculate the
optimal return value to be:
*

*
Thus
*

*
If we examine different values of ** we get different
optimal actions in **S*_{1}*.
*

For example:

*
Note that as ** increases the optimal policy at **S*_{1}*
changes from **a*_{12}* to **a*_{11}*.
*

*Yishay Mansour*

*1999-11-24*