Reviews: Learning and Testing Causal Models with Interventions

Neural Information Processing Systems 

The paper works on testing the interventional closeness of two causal models defined on the same causal graph. This is a non-trivial problem and authors make several key observations that may be useful for further research in the field, not only for testing, but also for learning causal graphs and causal models. This is by far the best paper in my batch and I would like to thank the authors for the well-written manuscript. Section 1.3, titled "Overview of our techniques" is especially well written and nicely summarizes the approach. There are some typos, which I believe the authors will fix in the camera ready if the paper is accepted.