Bayesian estimation
Bayesian estimation
•Goal is to choose the label (skin or ~skin) that maximizes the posterior
–this is called Maximum A Posteriori (MAP) estimation
likelihood
posterior (normalized)
0.5
•Suppose the prior is uniform:  P(skin) = P(~skin) =
= minimize probability of misclassification
–in this case                                          ,
–maximizing the posterior is equivalent to maximizing the likelihood
»                                                   if and only if 
–this is called Maximum Likelihood (ML) estimation