2bba9f4124283edd644799e0cecd45ca-AuthorFeedback.pdf

We thank all the reviewers for their constructive feedback. We address the key questions and concerns below. This is shown in Eq. 1 below. Therefore, this is not a valid counterexample to ρ -projection's handling of other forms of policy invariance. The ESOR values in Table 1 shows the number of iterations taken to reach expert's ESOR. However, they differ in the type of query used.