A MAP-based explanation is a most likely configuration of a specific set of variables given evidence. This type of explanations is more general and more difficult to compute, compared to MPE-based explanations [UAI02:P].

SamIam supports both approximate [UAI01:PD, UAI02:PD] and exact [UAI03:PD] algorithms for MAP-based explanations. SamIam is currently the only tool of this scope which provides a comprehensive engine for computing MAP-based explanations.