Show that the mean square estimate is the mean of the posterior den- sity. Show this by differentiating eq. (13) at the lecture slides with respect to and set the obtained gradient to zero. Now the conditional Bayes cost to be minimized is of the form B(|z) = (0|z)d (13)