Formal Privacy for Functional Data with Gaussian Perturbations
Authors: Ardalan Mirshani, Matthew Reimherr, Aleksandra Slavković
ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Simulations and an application to diffusion tensor imaging are briefly presented, with extensive additions included in a supplement. In Section 5 simulations highlight the role of different parameters, while Section 6 contains an application of Diffusion Tensor Imaging of Multiple Sclerosis patients. |
| Researcher Affiliation | Academia | 1Department of Statistics, Pennsylvania State University, State College, PA, USA. |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide concrete access to its source code. It mentions using data from an R package but does not state that their specific methodology's code is released. |
| Open Datasets | Yes | In this section we illustrate our method on an application involving brain scans (diffusion tensor imaging, DTI) that give fractional anisotropy (FA) tract profiles for the corpus callosum (CCA) and the right corticospinal tract (RCST) for patients with multiple sclerosis as well as controls; data are part of the refund (Huang et al., 2016) R package. |
| Dataset Splits | Yes | The first is regular Cross Validation, CV, and the second we call Private Cross Validation, PCV. In CV we fix φ and then take the ρ that gives the minimum 10-fold cross validation score. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments. |
| Software Dependencies | Yes | data are part of the refund (Huang et al., 2016) R package. |
| Experiment Setup | Yes | The mean function, sample size and DP parameters will also be set as µ(t) = 0.1 sin(πt), N = 25, (ϵ = 1, δ = 0.1), respectively. We vary the penalty, φ, from 10 6 to 1 to consider its effect. We varied φ in range [10 4, 0.1] for each of the kernels but ρ will be varied in [0.01, 0.1], [0.05, 0.5] and [0.2, 1] for C1,C3 and C4 respectively. |