NeurIPS 2020

Succinct and Robust Multi-Agent Communication With Temporal Message Control


Meta Review

with scores of (7, 5, 5, 7) this submission turned out to be difficult to adjudicate. All of the reviewers had some positive points to mention in line 2 (Strengths) such as "this is an important topic" and "I like the idea," while R3 argued against the setting. R3 also critiqued the use of heuristics that were not well justified. Correctness was favorable for R1, while R3 said the experiments did not fully support the claims. Clarity and Prior Art coverage were generally good. Reproducibility was also generally thought to be good, except for R1. I should also mention that the reviewers were also pleased to have good empirical results and ablation studies. Given the wide variety of opinions in the reviews in various topics, I am inclined to stick with the numbers provided by the reviewers. This means that the average score will be 6.0. This average score gives the paper a decent chance to be accepted, although acceptance is not at all guaranteed.