AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

Composable Sketches for Functions of Frequencies: Beyond the Worst Case

Edith Cohen, Ofir Geri, Rasmus Pagh

ICML 2020 · Learning theory and algorithms › Sample complexity · approach: Generalization and limits under data constraints · 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 ℓ2 sampling instead of ℓ1 sampling to emulate high-moment samplingSampling sketches · SO out, SO in, AOL, CAIDA, and UGRemulation overhead for 3rd and 10th moments: SO out: 124.30 -> 3.74 (3rd); 600.44 -> 18.04 (10th); SO in: 299.80 -> 4.12 (3rd); 677.56 -> 9.30 (10th); AOL: 34.81 -> Improved
Use ℓ2 sampling for universal emulation rather than ℓ1 or concave-sublinear samplingSampling sketches · SO out, SO in, AOL, CAIDA, and UGRuniversal emulation overhead: SO out: 624.58 vs 1.25 × 10^5 and 1672.60; SO in: 681.72 vs 4.20 × 10^4 and 1628.36; AOL: 92.92 vs 2.94 × 10^5 and 170.8Hurt
Use sampling by advice rather than ℓ1 PPSWOR for estimating high moments with small samplesSampling-by-advice sketch · AOL and Stack OverflowNRMSE for high frequency momentsMixed
Use with-replacement ℓ2 sampling rather than sampling by adviceHigh-moment estimation · AOL and Stack OverflowNRMSE for high frequency momentsImproved
Restrict inputs to distributions with an ℓq heavy hitterℓq sampling · Frequency distributions; target moments p ≥ qemulation overhead: 1/φImproved
Use ℓ1 sampling for the universal objective instead of ℓ2 samplingUniversal sampling · SO out, SO in, AOL, CAIDA, and UGRuniversal emulation overhead: SO out: 1.25 × 10^5 -> 624.58; SO in: 4.20 × 10^4 -> 681.72; AOL: 2.94 × 10^5 -> 92.92; CAIDA: 2.65 × 10^5 -> 301.28; UGImproved
Use multi-objective concave-sublinear sampling for universal emulation rather than ℓ2 samplingUniversal sampling · SO out, SO in, AOL, CAIDA, and UGRuniversal emulation overhead: SO out: 1.25 × 10^5 -> 1672.60; SO in: 4.20 × 10^4 -> 1628.36; AOL: 2.94 × 10^5 -> 170.84; CAIDA: 2.65 × 10^5 -> 846.15;Improved

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