AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Partitioned Learned Bloom Filters

Kapil Vaidya, Eric Knorr, Michael Mitzenmacher, Tim Kraska

ICLR 2021 · Learning theory and algorithms › Prediction and decision losses · approach: Adaptive testing and experiments · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use optimized score partitions and region-specific backup-filter rates instead of sandwichingLearned Bloom filter · Syntheticspace used at FPR 0.001: 8.8xImproved
Use optimized score partitions and region-specific backup-filter rates instead of sandwichingLearned Bloom filter · URLsspace used at FPR 0.001: 3.3xImproved
Use optimized score partitions and region-specific backup-filter rates instead of sandwichingLearned Bloom filter · EMBERspace used at FPR 0.001: 1.2xImproved
Use optimized score partitions and region-specific backup-filter rates instead of AdaBFLearned Bloom filter · Syntheticspace used at FPR 0.001: 6xImproved
Use optimized score partitions and region-specific backup-filter rates instead of AdaBFLearned Bloom filter · URLsspace used at FPR 0.001: 2.5xImproved
Use optimized score partitions and region-specific backup-filter rates instead of AdaBFLearned Bloom filter · EMBERspace used at FPR 0.001: 1.1xImproved
Increase the number of score partitions (ablation)PLBF · Synthetic, URLs, and EMBERspace saved versus optimal Bloom filterImproved
Increase model quality (ablation)Learned Bloom filter methods · Syntheticspace used at FPR 0.001Improved
Use KL divergence as the model-training loss instead of MSE (ablation)PLBF model · Model quality experimentspace usedNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy