BayesiaLab
Represent, evaluate and learn your action policies
BayesiaLab completes the decision aid brought by Bayesian networks thanks to:
Decision Nodes for the representation of
your actions
Quality tables associated to the actions to describe the policies
Utility Nodes for the valuation of the states
Reinforcement learning algorithms for the automatic learning
of policies that optimize the sum of the expected Utilities, for static as well
as dynamic Bayesian networks, and for exact as well as approximate inference
|
Action Policy Learning