Explanation on the Policy Tree Splitting Criteria

[yuyangstat]

Hi, thank you for making the policy tree and interpreter tools available! They are very useful!

In the meantime, I find it unclear as why you choose to use $\sum_i \sum_k g_{ik} e_{ki}$ as in the official doc. I have a few questions regarding this:

1. How is $g_{ik}$ obtained? Could you please point out where it is computed in the file tree_criterion.pyx?
2. Can you help explain the logic behind such a design?
3. And are there any reference papers I can reach out to for more information?
4. Similar to this, the logic behind the policy interpreter also remains unclear to me, and it would be super helpful if you can help clarify that.

Thanks!