NeurIPS 2020

Adaptive Learned Bloom Filter (Ada-BF): Efficient Utilization of the Classifier with Application to Real-Time Information Filtering on the Web

Meta Review

This paper provides an interesting new version of a bloom filter that takes a learning-based approach to find a useful set of practical wins in this important data structure / application setting. The reviewers appreciated the insightful approach, and I personally also found the authors responses to be helpful in considering the paper as well. The reviewers have come to a clear consensus that this paper is ready for NeurIPS, and I support acceptance -- with the clear expectation that the authors will use the reviewer feedback to further strengthen the paper. In particular, I think that addressing some of R4's questions that were well answered in the author response in the main paper text would be helpful to other readers as well.