NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:5909
Title:MaxGap Bandit: Adaptive Algorithms for Approximate Ranking


		
Although the problem is not properly motivated (some good examples in information retrieval and databases should be detailed), and despite the weaknesses of the results pointed in the reviews, the paper is clear and the originality of the work well established after the rebuttal. I thus recommend to accept it, and I encourage the authors to follow the suggestions of the reviews for the final version (if the paper is finally accepted) or before a re-submission (if it is not).