Inference

SamIam supports a number of algorithms for inference in Bayesian networks, including:

• Three implementations of the jointree algorithm based on: the Hugin architecture, the Shenoy-Shafer architecture, and a new architecture that combines the best of previous architectures [ESQARU03:PD]

• An implementation of the Recursive Conditioning algorithm with a time-space tradeoff engine [AIJ01:D, IJCAI03:AD]

SamIam supports non-classical queries, such as the computation of partial derivatives with respect to network parameters [NIPS02:PD]. It also supports query-specific inference which prunes the Bayesian network based on given query before inference algorithms are applied.