Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
133 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Bayesian Inference Under Differential Privacy With Bounded Data (2405.13801v2)

Published 22 May 2024 in stat.ME and cs.CR

Abstract: We describe Bayesian inference for the parameters of Gaussian models of bounded data protected by differential privacy. Using this setting, we demonstrate that analysts can and should take constraints imposed by the bounds into account when specifying prior distributions. Additionally, we provide theoretical and empirical results regarding what classes of default priors produce valid inference for a differentially private release in settings where substantial prior information is not available. We discuss how these results can be applied to Bayesian inference for regression with differentially private data.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (43)
  1. Hypothesis testing for differentially private linear regression. Advances in Neural Information Processing Systems, 35:14196–14209, 2022.
  2. Differentially private uniformly most powerful tests for binomial data. Advances in Neural Information Processing Systems, 31, 2018.
  3. Differentially private significance tests for regression coefficients. Journal of Computational and Graphical Statistics, 28(2):440–453, 2019.
  4. Differentially private Bayesian inference for exponential families. Advances in Neural Information Processing Systems, 31, 2018.
  5. Differentially private Bayesian linear regression. Advances in Neural Information Processing Systems, 32, 2019.
  6. Kenneth Alexander Brownlee. Statistical theory and methodology in science and engineering. https://people.sc.fsu.edu/~jburkardt/datasets/regression/x20.txt, 1965. Accessed on 4 May 2024.
  7. Differentially private nonparametric hypothesis testing. In Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security, pages 737–751, 2019.
  8. Probability matching priors. In D K Dey and C R Rao, editors, Bayesian Thinking, volume 25 of Handbook of Statistics, pages 91–114. Elsevier, 2005.
  9. Differential privacy for Bayesian inference through posterior sampling. Journal of Machine Learning Research, 18(11):1–39, 2017.
  10. Differential privacy for social science inference. Sloan Foundation Economics Research Paper, (2676160), 2015.
  11. Differentially private confidence intervals. arXiv preprint arXiv:2001.02285, 2020.
  12. Calibrating noise to sensitivity in private data analysis. In Theory of Cryptography: Third Theory of Cryptography Conference, TCC 2006, New York, NY, USA, March 4-7, 2006. Proceedings 3, pages 265–284. Springer, 2006.
  13. Parametric bootstrap for differentially private confidence intervals. In International Conference on Artificial Intelligence and Statistics, pages 1598–1618. PMLR, 2022.
  14. Andrew Gelman. Prior distributions for variance parameters in hierarchical models (comment on article by Browne and Draper). Bayesian Analysis, 1(3):515–534, 2006.
  15. Rényi differential privacy mechanisms for posterior sampling. Advances in Neural Information Processing Systems, 30, 2017.
  16. Ruobin Gong. Exact inference with approximate computation for differentially private data via perturbations. Journal of Privacy and Confidentiality, 12(2), 2022.
  17. Peter D Hoff. A First Course in Bayesian Statistical Methods, volume 580. Springer, 2009.
  18. Differentially private statistical inference through beta-divergence one posterior sampling. Advances in Neural Information Processing Systems, 36, 2024.
  19. Data augmentation mcmc for Bayesian inference from privatized data. Advances in neural information processing systems, 35:12732–12743, 2022.
  20. Finite sample differentially private confidence intervals. arXiv preprint arXiv:1711.03908, 2017.
  21. On the’semantics’ of differential privacy: A Bayesian formulation. Journal of Privacy and Confidentiality, 6(1), 2014.
  22. Prior-itizing privacy: A Bayesian approach to setting the privacy budget in differential privacy. arXiv preprint arXiv:2306.13214, 2023.
  23. Assessing statistical disclosure risk for differentially private, hierarchical count data, with application to the 2020 us decennial census. Statistica Sinica, 34:to appear, 2024.
  24. The test of tests: A framework for differentially private hypothesis testing. In International Conference on Machine Learning, pages 16131–16151. PMLR, 2023.
  25. Bayesian and frequentist semantics for common variations of differential privacy: Applications to the 2020 census. arXiv preprint arXiv:2209.03310, 2022.
  26. Pufferfish: A framework for mathematical privacy definitions. ACM Transactions on Database Systems (TODS), 39(1):1–36, 2014.
  27. John K Kruschke. Bayesian approaches to testing a point (“null”) hypothesis. In John K Kruschke, editor, Doing Bayesian Data Analysis (Second Edition), pages 335–358. Academic Press, Boston, second edition edition, 2015.
  28. Bayesian inference for estimating subset proportions using differentially private counts. Journal of Survey Statistics and Methodology, 10(3):785–803, 2022.
  29. Differential privacy and statistical disclosure risk measures: An investigation with binary synthetic data. Trans. Data Priv., 5(3):535–552, 2012.
  30. HDInterval: Highest (Posterior) Density Intervals, 2020. R package version 0.2.2.
  31. Differential privacy without sensitivity. Advances in Neural Information Processing Systems, 29, 2016.
  32. Study of lead exposure from automobile exhaust as a risk for nephrotoxicity among traffic policemen. American Journal of Nephrology, 21(4):274–279, 2001.
  33. truncdist: Truncated Random Variables, 2016. R package version 1.0-2.
  34. The Bayesian lasso. Journal of the American Statistical Association, 103(482):681–686, 2008.
  35. Differentially private hypothesis testing with the subsampled and aggregated randomized response mechanism. Statistica Sinica, 34:to appear, 2024.
  36. Locally private Bayesian inference for count models. In International Conference on Machine Learning, pages 5638–5648. PMLR, 2019.
  37. Or Sheffet. Differentially private ordinary least squares. In International Conference on Machine Learning, pages 3105–3114. PMLR, 2017.
  38. Objective Bayesian analysis for the multivariate normal model. Bayesian Statistics, 8:525–562, 2007.
  39. Yu-Xiang Wang. Revisiting differentially private linear regression: optimal and adaptive prediction & estimation in unbounded domain. arXiv preprint arXiv:1803.02596, 2018.
  40. Privacy for free: Posterior sampling and stochastic gradient Monte Carlo. In International Conference on Machine Learning, pages 2493–2502. PMLR, 2015.
  41. Functional mechanism: regression analysis under differential privacy. Proceedings of the VLDB Endowment, 5(11):1364–1375, 2012.
  42. Dp-fast mh: Private, fast, and accurate Metropolis-Hastings for large-scale Bayesian inference. In International Conference on Machine Learning, pages 41847–41860. PMLR, 2023.
  43. On the differential privacy of Bayesian inference. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 30, 2016.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets