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.