NeurIPS 2020

Trust the Model When It Is Confident: Masked Model-based Actor-Critic

Meta Review

The reviewers generally appreciated the paper, although Reviewer #1 raised the concern that there are some nontrivial failure cases in the algorithm, which are not explained by the theory. More generally, the reviewers were concerned about a wide theory/implementation gap. The author response clarified some of how this could be accomplished. The authors are encouraged to spend some time clarifying this issue.