NeurIPS 2020

Online Algorithm for Unsupervised Sequential Selection with Contextual Information

Meta Review

While the reviewers find the paper interesting and the results correct, there were some concerns regarding the novelty of the techniques beyond Verma et al AISTATS 2019 and GLM bandit literature. I am only partially convinced by the author's claim of novelty in their response. While the contextual setting is novel in unsupervised sequential selection, the techniques used for this extension under GLM and CWD assumptions seem incremental. However, the problem studied seems important and the results are new. Therefore, I am open to accepting this paper.