Bayesian Networks
nodes = variables
edges = direct influence
Graph structure encodes independence assumptions:
 Letter  conditionally independent of Intelligence given Grade
A
B
C
CPD P(G|D,I)
Letter
Grade
SAT
Intelligence
Difficulty
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