Parameter Adjustment in Bayes Networks. The generalized noisy OR-gate
This work addresses incremental improvements in Bayesian network methods for probabilistic reasoning, with potential applications in domains like AI and decision-making.
The paper tackles the problem of learning and inference in Bayesian networks by introducing a new parameter distribution model based on Gaussian functions and generalizing the noisy OR-gate for multivalued variables, resulting in an algorithm that computes probabilities in time proportional to the number of parents, even in networks with loops.
Spiegelhalter and Lauritzen [15] studied sequential learning in Bayesian networks and proposed three models for the representation of conditional probabilities. A forth model, shown here, assumes that the parameter distribution is given by a product of Gaussian functions and updates them from the _ and _r messages of evidence propagation. We also generalize the noisy OR-gate for multivalued variables, develop the algorithm to compute probability in time proportional to the number of parents (even in networks with loops) and apply the learning model to this gate.