Differentially Private Sampling from Distributions

Authors: Sofya Raskhodnikova, Satchit Sivakumar, Adam Smith, Marika Swanberg

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We provide tight upper and lower bounds for the dataset size needed for this task for three natural families of distributions: arbitrary distributions on {1, . . . , k}, arbitrary product distributions on {0, 1}d, and product distributions on on {0, 1}d with bias in each coordinate bounded away from 0 and 1. We demonstrate that, in some parameter regimes, private sampling requires asymptotically fewer observations than learning a description of P nonprivately; in other regimes, however, private sampling proves to be as difficult as private learning.
Researcher Affiliation Academia Sofya Raskhodnikova Department of Computer Science Boston University sofya@bu.edu Satchit Sivakumar Department of Computer Science Boston University satchit@bu.edu Adam Smith Department of Computer Science Boston University ads22@bu.edu Marika Swanberg Department of Computer Science Boston University marikas@bu.edu
Pseudocode No The paper describes algorithms and proof techniques in prose but does not provide structured pseudocode or clearly labeled algorithm blocks.
Open Source Code No The paper does not contain any statement about releasing source code for the methodology, nor does it provide a link to a code repository.
Open Datasets No The paper is theoretical and focuses on mathematical proofs and bounds for distribution classes (e.g., 'distributions on {1, . . . , k}'), rather than conducting experiments on specific, named datasets.
Dataset Splits No The paper is theoretical and does not describe empirical experiments that would involve dataset splits for training, validation, or testing.
Hardware Specification No The paper is theoretical and does not describe empirical experiments, therefore no hardware specifications for running experiments are mentioned.
Software Dependencies No The paper is theoretical and does not describe empirical experiments, therefore no specific software dependencies with version numbers are mentioned.
Experiment Setup No The paper is theoretical and does not describe empirical experiments with hyperparameters or system-level training settings.