Dynamic Weighting A* Search-based MAP Algorithm for Bayesian Networks

Xiaoxun Sun, Marek J. Druzdzel, Changhe Yuan

In this paper we propose the Dynamic Weighting A* (DWA*) search algorithm for solving MAP problems in Bayesian networks. By exploiting asymmetries in the distribution of MAP variables, the algorithm is able to greatly reduce the search space and offer excellent performance both in terms of accuracy and efficiency.