Bayesian policy selection using active inference