Optimal Differential Privacy Composition for Exponential Mechanisms
Authors: Jinshuo Dong, David Durfee, Ryan Rogers
ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We present plots of our results in Figure 1 for the homogeneous case, plotting εg as a function of k. |
| Researcher Affiliation | Collaboration | 1Applied Mathematics and Computational Sciences, University of Pennsylvania 2Data Science Applied Research, Linked In. |
| Pseudocode | No | The paper describes recursive formulas and algorithms in prose but does not include any explicitly labeled 'Pseudocode' or 'Algorithm' block. |
| Open Source Code | No | The paper does not provide any explicit statements about releasing source code or links to a code repository for the methodology described. |
| Open Datasets | No | The paper is theoretical/mathematical, focusing on composition bounds. It does not describe experiments involving training on datasets; the 'results' in Figure 1 are plots of derived bounds. |
| Dataset Splits | No | The paper focuses on theoretical bounds and numerical comparisons of these bounds. It does not describe experiments that would require dataset splits like training, validation, or test sets. |
| Hardware Specification | No | The paper does not mention any specific hardware used for its computations or for generating the plots shown, such as CPU/GPU models or cloud resources. |
| Software Dependencies | No | The paper does not specify any software dependencies with version numbers, such as programming languages, libraries, or solvers used for computations. |
| Experiment Setup | No | The paper is theoretical and presents numerical comparisons of derived bounds. It does not describe an experimental setup with hyperparameters, training configurations, or other system-level settings typically found in empirical studies. |