<< Click to Display Table of Contents >> Navigation:  Building blocks of GeNIe > Inference algorithms > Bayesian networks algorithms > Stochastic sampling algorithms > Backward Sampling

The Backward Sampling algorithm is described in (Fung & del Favero 1994).

It attempts to defy the problems with unlikely evidence by sampling backward from the evidence nodes. As nodes can be sampled both backward and forward, depending on whether they have direct ancestors or descendants sampled, this algorithms is an ingenious extension to forward sampling algorithms.