EBNN Example

Combining Inductive and Analytical Learning

There is one network for each of the Horn clauses in the domain theory.
EBNN uses the top network to calculate the partial derivative of the prediction with respect to each feature of the instance. (i.e., how much does the output change as I tweak BottomIsFlat?).
These derivatives are given to the bottom network which is trained with a variation of TangentProp.