by Leslie Tung
Context specific independence can provide compact representation of the conditional probabilities in Bayesian networks, when some variables are only relevant in specific contexts. We present an algorithm that exploits context specific independence in clique tree propagation. The algorithm is based on a query based contextual variable elimination algorithm that eliminates in turn the variables not needed in an answer. We extend this algorithm to producing the posterior on all variables efficiently and allow the incremental addition of evidence. We present empirical results showing that this new algorithm is efficient, both in time and in space (as compared to the Hugin architecture), on computing posterior probabilities for Bayesian networks that exhibit context specific independence.