Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
This paper makes some significant contributions to the problem of policy evaluation in the presence of unobserved confounders (assuming access to some proxy variables). The theoretical and technical contributions of this paper are pretty significant. There have been some inputs from reviewers regarding improving the experimental setup of this paper by utilizing more real-world high dimensional data. Based on reviewer feedback and rebuttals, I recommend this paper to be accepted.