Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Learning-augmented count-min sketches via Bayesian nonparametrics
Emanuele Dolera, Stefano Favaro, Stefano Peluchetti
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use CMS-PYP instead of CMS-DP | count-min sketch point queries · Zipf synthetic streams, bin (0,256] | MAE | Improved |
| use CMS-PYP instead of CMS-DP as tail becomes heavier (smaller Zipf exponent c) | count-min sketch point queries · Zipf synthetic streams | MAE | Improved |
| use CMS-PYP instead of CMS-NIGP | count-min sketch point queries · Zipf synthetic streams, c=1.3 and 1.6 (J=320, N=2) | MAE | Improved |
| use CMS-PYP instead of CMS-NIGP with more hash functions | count-min sketch point queries · Zipf synthetic streams, c=1.3 and 1.6 (J=160, N=4) | MAE | No clear effect |
| use CMS-PYP instead of CMM and BDCM | count-min sketch point queries · Zipf synthetic streams, low-frequency tokens | MAE | Improved |
| use CMS-PYP instead of CMS-DP | count-min sketch point queries · 20 Newsgroups and Enron, low-frequency tokens | MAE | Improved |
| use CMS-PYP instead of CMM | count-min sketch point queries · 20 Newsgroups and Enron, low-frequency tokens | MAE | No clear effect |
| use CMS-PYP instead of CMS-DP or CMM | count-min sketch point queries · WikiText-103 and 1BWLMB | MAE | No clear effect |
Closest papers by meaning
- Recursive PAC-Bayes: A Frequentist Approach to Sequential Prior Updates with No Information Loss (NeurIPS 2024 · similarity 0.84)
- A Bayesian Approach to Data Point Selection (NeurIPS 2024 · similarity 0.82)
- Scalable Bayesian Learning with posteriors (ICLR 2025 · similarity 0.82)
- Model-Based Minimum Bayes Risk Decoding for Text Generation (ICML 2024 · similarity 0.82)
- State-of-art minibatches via novel DPP kernels: discretization, wavelets, and rough objectives (NeurIPS 2026 · similarity 0.82)
- MASS: Mathematical Data Selection via Skill Graphs for Pretraining Large Language Models (ICML 2025 · similarity 0.82)
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.
