Dynamic Bayesian Network Action Set

Bayesian networks consist of variables, which are represented as nodes, and relationships among these variables, which are represented as directed edges. The variables are a mix of hidden and observed variables. The dynamic Bayesian network (dynBnet) action in the Dynamic Bayesian Network (dynamicBayesianNetwork) action set computes probabilistic distributions of hidden variables, given the values of observed variables. This action is referred to as dynamic because it is used to update the distributions of hidden variables as new values of observed variables become available over time. Arranging variables in a network provides significant computational savings, allowing solutions to problems that could not be tackled otherwise.

This section describes the overall methodology of dynamic Bayesian networks that the dynBnet action implements.