Rényi Pufferfish Privacy: General Additive Noise Mechanisms and Privacy Amplification by Iteration (2312.13985v2)
Abstract: Pufferfish privacy is a flexible generalization of differential privacy that allows to model arbitrary secrets and adversary's prior knowledge about the data. Unfortunately, designing general and tractable Pufferfish mechanisms that do not compromise utility is challenging. Furthermore, this framework does not provide the composition guarantees needed for a direct use in iterative machine learning algorithms. To mitigate these issues, we introduce a R\'enyi divergence-based variant of Pufferfish and show that it allows us to extend the applicability of the Pufferfish framework. We first generalize the Wasserstein mechanism to cover a wide range of noise distributions and introduce several ways to improve its utility. We also derive stronger guarantees against out-of-distribution adversaries. Finally, as an alternative to composition, we prove privacy amplification results for contractive noisy iterations and showcase the first use of Pufferfish in private convex optimization. A common ingredient underlying our results is the use and extension of shift reduction lemmas.
- Deep learning with differential privacy. In Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security, CCS ’16, page 308–318.
- Faster high-accuracy log-concave sampling via algorithmic warm starts. arXiv:2302.10249.
- Privacy of noisy stochastic gradient descent: More iterations without more privacy loss. In NeurIPS.
- Concentrated differential privacy: Simplifications, extensions, and lower bounds. In Theory of Cryptography, pages 635–658. Springer Berlin Heidelberg.
- Protecting global properties of datasets with distribution privacy mechanisms. In Proceedings of The 26th International Conference on Artificial Intelligence and Statistics, volume 206 of Proceedings of Machine Learning Research, pages 7472–7491. PMLR.
- Ding, N. (2022). Kantorovich mechanism for pufferfish privacy. In Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, volume 151 of Proceedings of Machine Learning Research, pages 5084–5103. PMLR.
- The algorithmic foundations of differential privacy. Foundations and Trends in Theoretical Computer Science, 9(3–4):211–407.
- Privacy amplification by iteration. 2018 IEEE 59th Annual Symposium on Foundations of Computer Science (FOCS), pages 521–532.
- Beyond worst-case analysis in private singular vector computation. In Proceedings of the Forty-Fifth Annual ACM Symposium on Theory of Computing, STOC ’13, page 331–340.
- Blowfish privacy: Tuning privacy-utility trade-offs using policies. In Proceedings of the 2014 ACM SIGMOD International Conference on Management of Data, SIGMOD ’14, page 1447–1458. Association for Computing Machinery.
- Investigating membership inference attacks under data dependencies. In 2023 2023 IEEE 36th Computer Security Foundations Symposium (CSF) (CSF), pages 194–209. IEEE Computer Society.
- Local obfuscation mechanisms for hiding probability distributions. In Computer Security – ESORICS 2019, pages 128–148, Cham. Springer International Publishing.
- Deploying and evaluating pufferfish privacy for smart meter data. In 2015 IEEE 12th Intl Conf on Ubiquitous Intelligence and Computing and 2015 IEEE 12th Intl Conf on Autonomic and Trusted Computing and 2015 IEEE 15th Intl Conf on Scalable Computing and Communications and Its Associated Workshops (UIC-ATC-ScalCom), pages 229–238.
- Pufferfish: A framework for mathematical privacy definitions. ACM Transactions on Database Systems, 39(1).
- On the loss landscape of adversarial training: Identifying challenges and how to overcome them. In Advances in Neural Information Processing Systems, pages 21476–21487.
- Mironov, I. (2017). Rényi differential privacy. In 2017 IEEE 30th Computer Security Foundations Symposium (CSF), pages 263–275.
- Making big money from small sensors: Trading time-series data under pufferfish privacy. In IEEE INFOCOM 2019 - IEEE Conference on Computer Communications, pages 568–576.
- An optimal pufferfish privacy mechanism for temporally correlated trajectories. IEEE Access, 6:37150–37165.
- Rider, P. R. (1957). Generalized cauchy distributions. Annals of the Institute of Statistical Mathematics, 9:215–223.
- Pufferfish privacy mechanisms for correlated data. In Proceedings of the 2017 ACM International Conference on Management of Data, SIGMOD ’17, page 1291–1306.
- Verdú, S. (2023). The cauchy distribution in information theory. Entropy, 25(2).