Cascaded Learned Bloom filter for Optimal Model-Filter Size Balance and Fast Rejection

Atsuki Sato, Yusuke Matsui

Keywords: learned Bloom filter, learned index, membership query, optimization, dynamic programming

TL;DR: We propose a novel learned Bloom filter with a cascaded structure that optimizes model-filter size balance and minimizes rejection time, achieving up to 24% memory savings and 14x faster rejection than the state-of-the-art method.

Abstract: Recent studies have demonstrated that learned Bloom filters, which combine machine learning with the classical Bloom filter, can achieve superior memory efficiency. However, existing learned Bloom filters face two critical unresolved challenges: the balance between the machine learning model size and the Bloom filter size is not optimal, and the reject time cannot be minimized effectively. We propose the Cascaded Learned Bloom Filter (CLBF) to address these issues. Our optimization approach based on dynamic programming automatically selects configurations that achieve an optimal balance between the model and filter sizes while minimizing reject time. Experiments with real-world datasets show that CLBF reduces memory usage by up to 24% and decreases reject time by up to 14 times compared to the state-of-the-art learned Bloom filter.

Primary Area: optimization

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 11055