arXiv 26 Jul 2022 · Statistics — Machine Learning
arXiv:2207.12602 · PDF · DOI · OpenAlex · Extracted main text
Constructing a differentially private (DP) estimator requires deriving the maximum influence of an observation, which can be difficult in the absence of exogenous bounds on the input data or the estimator, especially in high dimensional settings. This paper shows that standard notions of statistical depth, i.e., halfspace depth and regression depth, are particularly advantageous in this regard, both in the sense that the maximum influence of a single observation is easy to analyze and that this value is typically low. This is used to motivate new approximate DP location and regression estimators using the maximizers of these two notions of statistical depth. A more computationally efficient variant of the approximate DP regression estimator is also provided. Also, to avoid requiring that users specify a priori bounds on the estimates and/or the observations, variants of these DP mechanisms are described that satisfy random differential privacy (RDP), which is a relaxation of differential privacy provided by Hall, Wasserman, and Rinaldo (2013). We also provide simulations of the two DP regression methods proposed here. The proposed estimators appear to perform favorably relative to the existing DP regression methods we consider in these simulations when either the sample size is at least 100-200 or the privacy-loss budget is sufficiently high.
appendix boundary found by appendix_command · 90% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Nissim, K., Raskhodnikova, S., and Smith, A (2007) Smooth sensitivity and sampling in private data analysis | 1.000 | 23 | 4 | 100% |
| 2 | Rousseeuw, P. J. and Hubert, M (1999) Regression depth | 1.000 | 9 | 3 | 100% |
| 3 | Dwork, C., McSherry, F., Nissim, K., and Smith, A (2006) Calibrating noise to sensitivity in private data analysis | 1.000 | 7 | 3 | 100% |
| 4 | Chen, Y., Machanavajjhala, A., Reiter, J. P., and Barrientos, A. F (2016) Differentially private regression diagnostics | 1.000 | 6 | 3 | 100% |
| 5 | Hall, R., Wasserman, L., and Rinaldo, A (2013) Random differential privacy | 0.928 | 4 | 3 | 100% |
| 6 | Donoho, D. L. and Gasko, M (1992) Breakdown properties of location estimates based on halfspace depth and projected outlyingness | 0.874 | 9 | 2 | 100% |
| 7 | Mizera, I (2002) On depth and deep points: a calculus | 0.874 | 5 | 2 | 100% |
| 8 | Tukey, J. W (1975) Mathematics and the picturing of data | 0.811 | 4 | 2 | 100% |
| 9 | Van Aelst, S., Rousseeuw, P. J., Hubert, M., and Struyf, A (2002) The deepest regression method | 0.737 | 3 | 2 | 100% |
| 10 | Karwa, V. and Vadhan, S (2018) Finite sample differentially private confidence intervals | 0.693 | 6 | 1 | 100% |
Showing the top 10 of 59 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.
| Citing paper | Intensity | Mentions | Sections | |
|---|---|---|---|---|
| 1 | Differentially Private Two-Stage Gradient Descent for Instrumental Variable Regression | 0.405 | 1 | 1 |