I am trying to learn a model from a set of positive examples.
A model being considered as made of several constraints, it looks much
like a system of equations.
I would like to represent each individual with several trees, each tree
being the right inside of the equations forming the system.

I guess using ADF is a good solution, but, I am not sure that is the best
way to proceed... Is there any better way to proceed ?

I have been implementing a simplified version of the problem with only one
tree, which follows the example of tutorial 4. The whole thing work well
to find a single equation.
Now, what should I do to transform an example of regression such as
tutorial 4 into something that finds let's say, a system of 3 equations
with three variables ?

Any suggestion would be greatly appreciated...