aa36c88c27650af3b9868b723ae15dfc-AuthorFeedback.pdf

We thank all reviewers for their time and valuable comments. We thank this reviewer for the positive feedback! "The theoretical sample complexity is not significantly improved over previously-known methods." The main contribution of our paper is to show that an existing and popular algorithm (i.e., group-sparse regularized We view the sample complexity improvement over the dependence on k as a side benefit of our analysis. "It would be interesting to see a more thorough empirical evaluation, to compare with the interaction screening The main contribution of our paper is theoretical. Our graph has diamond shape (Figure 1 of our paper), 10 variables and edge weight 0.2. This observation is actually the starting point of our paper. We will include this discussion in our paper. "The presentation is quite technical...the Ising case seems to be enough to introduce the main idea...but a lot of For learning non-binary graphical models, we see a benefit of using the group-sparse (i.e., the l "Experiments are only presented for rather small examples (up to 14 variables, up to k = 6)."