477bdb55b231264bb53a7942fd84254d-AuthorFeedback.pdf

Unsurprisingly, this approach performed poorly. Interpretable policy's dependence on demonstrator identity - Discretized PNTs learn splitting criterion dependent Please see Figure 2 in the supplementary for a depiction. Rephrasing Hypothesis - We thank R3 for these suggestions. "Interpretability" is vaguely defined - We agree that our study left the definition of "interpretability" open to the Details of synthetic schedule generation - This information is located within the supplementary. However, we agree this would be an interesting metric to pursue in future work.