We will further try to simplify the paper in general.

We genuinely appreciate all three reviewers' (#1,#2,#3,#4) valuable suggestions to strengthen our paper. More details are referred to Reviewer #3's Q2. We sincerely appreciate your suggestion and will revise the caption in figure 4 for better readability. Reply: It is a great observation. Thus, IL appears to be a main contributor for L2O generalizing across different optimizees.

We have addressed the reviewers' comments by running seven new experiments, which shed useful new light on some of R2: GSP seems intuitively dependent on parametrization, can you discuss? R3: Does the benefit of aggregation disappear once you take into account the number of responses required? R3: How do the experimenters avoid subjects merely making the same response 10 times? R3: It would be worth discussing how the technique differs from e.g. GSP is more mode-seeking than MCMCP, but nonetheless recovers the utility function more reliably (Fig. D).

We thank all four reviewers for their constructive comments. We respond to each reviewer's comments separately below. We also report some results from new experiments suggested by the reviewers. This idea is illustrated in Figure 2 in the paper. Second, we followed the reviewer's suggestion and developed a temporal version of MoCo as a new baseline: MoCo-Img: 46.6% in the labeled S dataset).

We thank all three reviewers for their constructive and valuable feedback. They found our paper to be a "very interesting We do not have a final answer yet, but we will discuss some hypotheses in the final version. We will include these results in supplementary material. R1: Consider testing semantically relevant perturbations. Experimentally, mice allow for genetic tools for large scale recordings ( 8000 units).