Differentially Private Anonymized Histograms

Authors: Ananda Theertha Suresh

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Motivated by these applications, we propose the first differentially private mechanism to release anonymized histograms that achieves near-optimal privacy utility trade-off both in terms of number of items and the privacy parameter. Further, if the underlying histogram is given in a compact format, the proposed algorithm runs in time sub-linear in the number of items. For anonymized histograms generated from unknown discrete distributions, we show that the released histogram can be directly used for estimating symmetric properties of the underlying distribution. The paper presents theoretical guarantees and algorithms (PRIVHIST) but does not include empirical studies with actual data, performance metrics, or comparisons on datasets.
Researcher Affiliation Industry Ananda Theertha Suresh Google Research, New York theertha@google.com
Pseudocode Yes Algorithm PRIVHIST Input: anonymized histogram h in terms of prevalences i.e., {(r, 'r) : 'r > 0}, privacy cost . Parameters: 1 = 2 = 3 = /3. Output: DP anonymized histogram H and N (an estimate of n). ... Algorithm PRIVHIST-LOWPRIVACY ... Algorithm PRIVHIST-HIGHPRIVACY
Open Source Code No The paper does not provide any explicit statement about making the source code available, nor does it include links to a code repository.
Open Datasets No The paper is theoretical and focuses on algorithm design and theoretical properties. It does not conduct empirical experiments with datasets; therefore, it does not mention public or open datasets for training.
Dataset Splits No The paper is theoretical and does not describe any empirical experiments or dataset splits for training, validation, or testing.
Hardware Specification No The paper is theoretical and does not describe any empirical experiments that would require specific hardware. Therefore, no hardware specifications are mentioned.
Software Dependencies No The paper is theoretical and focuses on algorithm design and theoretical properties. It does not describe any empirical experiments that would require specific software dependencies with version numbers.
Experiment Setup No The paper is theoretical and does not describe any empirical experiments. Therefore, no experimental setup details like hyperparameters or training configurations are provided.