Letterconditionally independent of Intelligence given Grade
One of the key benefits of
the propositional representation is our ability to represent our knowledge
without explicit enumeration of the worlds.In turns out that we can do the same in the probabilistic framework.The key idea, introduced by Pearl in the
Bayesian network framework, is to use locality of interaction.We represent the interactions of the
variables using a graph structure.For
each variable, we select as parents the variables that
directly influence it.
This is an assumption which
seems to be a fairly good approximation of the world in many cases.
Each cpt is independent of
others – changing values locally does not affect coherence of the model