Improving Sparse Vector Technique with Renyi Differential Privacy
Authors: Yuqing Zhu, Yu-Xiang Wang
NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive empirical evaluation demonstrates the merits of Gaussian SVT over the Laplace SVT and other alternatives, which encouragingly suggests that using Gaussian SVT as a drop-in replacement could make SVT-based algorithms more practical in downstream tasks. |
| Researcher Affiliation | Academia | Yuqing Zhu Department of Computer Science UC Santa Barbara CA 93106 yuqingzhu@ucsb.edu Yu-Xiang Wang Department of Computer Science UC Santa Barbara CA 93106 yuxiangw@cs.ucsb.edu |
| Pseudocode | Yes | Algorithm 1 Standard SVT |
| Open Source Code | No | The paper does not provide an explicit statement or link for open-source code for the methodology described. |
| Open Datasets | Yes | Exp. 3 (Real life data) We investigate various private screening methods with a realistic sequence of queries from running a k NN-based private-query release on the CIFAR-10 dataset. |
| Dataset Splits | No | The paper mentions using the CIFAR-10 dataset but does not specify the exact train/validation/test split percentages, sample counts, or methodology for creating these splits for reproducibility. |
| Hardware Specification | No | The paper does not provide specific details regarding the hardware (e.g., GPU/CPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers (e.g., programming language versions, library versions, or solver versions). |
| Experiment Setup | Yes | In this section, we conduct extensive numerical experiments to illustrate the behaviors of SVT variants. We will have three sets of experiments. ... The ratio between the query noise and the threshold noise is fixed λ1 = 2. ... (a) T = 100, c = 20, δ = 10 6 ... (b) T = 700, c = 20, δ = 10 10 ... The margin T = 1000 and σ1 = 210. The standard deviation of Gaussian and Laplace are aligned to be comparable. ... for δ = 10 6. |