Order-Invariant Cardinality Estimators Are Differentially Private

Authors: Charlie Dickens, Justin Thaler, Daniel Ting

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our analysis applies to essentially all popular cardinality estimation algorithms, and substantially generalizes and tightens privacy bounds from earlier works. Our approach is faster and exhibits a better utility-space tradeoff than prior art. ... We provide two experiments highlighting the practical benefits of our approach.
Researcher Affiliation Collaboration Charlie Dickens Yahoo charlie.dickens@yahooinc.com Justin Thaler Georgetown University justin.thaler@georgetown.edu Daniel Ting Meta dting@meta.com
Pseudocode Yes Algorithms 1: Differentially private cardinality estimation algorithms from black box sketches. ... Pseudocode for the modified algorithm... is given in Algorithm 1c.
Open Source Code Yes A small repository containing the experimental scripts and figure plotting has been provided.
Open Datasets No The paper describes abstract 'streams of samples' and does not refer to a specific named dataset, provide a link, or cite a source for public access.
Dataset Splits No The paper does not mention any validation dataset splits.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments, only noting 'Wall-clock update time' measurements.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., Python 3.x, PyTorch 1.x) needed to replicate the experiments.
Experiment Setup Yes Details of the experimental setup are in Appendix D. For Figure 1a, we simulate the algorithms (HLL, PHLL, QLL) for 210 updates for k {27, 28, . . . 212} buckets. ... For Figure 1b, we set the input cardinality at n = 220 and the privacy budget at ϵ = ln(2). We varied k {27, 28, . . . 212} and simulate the ϵ-DP methods, PHLL and QLL [22].