NeurIPS 2020

High-recall causal discovery for autocorrelated time series with latent confounders


Meta Review

The reviewers appreciated the theoretical and empirical improvements, and notably the clarity of the authors in stating the method and its uses: the method was not overstated and did not make outrageous claims, things that could be more common in current ML literature. The rebuttal improved the reviewer scores. They agreed on the importance of a real data example, and if the authors are able to incorporate this among their other suggestions it will make the paper stronger. I vote to accept.