Differentially Private Covariance Estimation
Authors: Kareem Amin, Travis Dick, Alex Kulesza, Andres Munoz, Sergei Vassilvitskii
NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our empirical results demonstrate lower reconstruction error for our algorithm when compared to other methods on both simulated and real-world datasets. and Finally, we perform an empirical evaluation of our algorithm, comparing it to existing methods on both synthetic and real-world datasets (Section 4). |
| Researcher Affiliation | Collaboration | Kareem Amin kamin@google.com Google Research NY Travis Dick tdick@cs.cmu.edu Carnegie Mellon University Alex Kulesza kulesza@google.com Google Research NY Andr es Mu noz Medina ammedina@google.com Google Research NY Sergei Vassilvitskii sergeiv@google.com Google Research NY |
| Pseudocode | Yes | Pseudocode for our method is given in Algorithm 1. and Pseudocode for their method is given in Algorithm 2 in the appendix. |
| Open Source Code | No | The paper does not provide any explicit statements or links for open-source code for the described methodology. |
| Open Datasets | Yes | We measure the performance of our algorithm on three different datasets: Wine, Adult, and Airfoil from the UCI repository2, These datasets have dimensions ranging from 13 to 108, and number of points from 200 to 49,000. 2https://archive.ics.uci.edu/ml/datasets/ |
| Dataset Splits | No | The paper uses datasets from the UCI repository but does not explicitly provide details about train/validation/test splits, proportions, or specific methods for creating these splits. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers (e.g., library names with versions) needed to replicate the experiments. |
| Experiment Setup | Yes | We run each algorithm with privacy parameter 2 {0.01, 0.1, 0.2, 0.5, 1.0, 2.0, 4.0}. For the Gaussian mechanism, we also varied the parameter δ 2 {1e 16, 1e 10, 1e 3} We ran each experiment 50 times, showing the average error in Figure 1. |