Document Open Access Logo

Optimal Partition Selection with Rényi Differential Privacy

Authors Charlie Harrison , Pasin Manurangsi

PDF

File

Thumbnail PDF
PDF
LIPIcs.FORC.2026.16.pdf
  • Filesize: 1.99 MB
  • 22 pages

Document Identifiers

Subject Classification

ACM Subject Classification
  • Security and privacy
  • Theory of computation → Design and analysis of algorithms
Keywords
  • Differentially Privacy
  • Partition Selection
  • Renyi Differentially Privacy

Metrics

Abstract

A common problem in private data analysis is the partition selection problem, where each user holds a set of partitions (e.g. keys in a GROUP BY operation) from a possibly unbounded set. The challenge here is in maximizing the set of released partitions while respecting a differential privacy constraint. Previous work [Desfontaines et al., 2021] presented an optimal (ε, δ)-DP algorithm when each user submits only a single partition. We generalize this approach to find the optimal algorithm under δ-approximate (α, ε)-Rényi differential privacy (RDP), which allows much tighter analysis under composition. Motivated by the non-existence of a general optimality result in the case where users submit multiple partitions each, we present an extension of our optimal algorithm tuned for L² bounded weighted partition selection which can be used as a drop-in improvement over the Gaussian mechanism any time the partition frequency is not also needed. We show that our primitive can be easily plugged into state of the art partition selection algorithms (PolicyGaussian from [Gopi et al., 2020] and MAD2R from [Justin Y. Chen et al., 2025]), improving performance both for parallel and sequential adaptive algorithms. Finally, we show that there is an inherent cost to algorithms which do support releasing the frequency as well as the partitions. Specifically, we formulate a basic notion of optimal approximate RDP algorithm for partition selection using additive noise, and show that there is a numerical separation between additive and non-additive noise mechanisms for this problem.

Cite As Get BibTex

Charlie Harrison and Pasin Manurangsi. Optimal Partition Selection with Rényi Differential Privacy. In 7th Symposium on Foundations of Responsible Computing (FORC 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 368, pp. 16:1-16:22, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026) https://doi.org/10.4230/LIPIcs.FORC.2026.16

Author Details

Charlie Harrison
  • Google, Austin, TX, USA
Pasin Manurangsi
  • Google Research, Bangkok, Thailand

References

  1. Miguel Aenlle. Daily financial news for 6000+ stocks. URL: https://www.kaggle.com/miguelaenlle/datasets.
  2. Kareem Amin, Jennifer Gillenwater, Matthew Joseph, Alex Kulesza, and Sergei Vassilvitskii. Plume: Differential privacy at scale. CoRR, abs/2201.11603, 2022. URL: https://arxiv.org/abs/2201.11603.
  3. Stuart Axelbrooke. Customer support on twitter, 2017. URL: https://doi.org/10.34740/KAGGLE/DSV/8841.
  4. James Bell, Adrià Gascón, Badih Ghazi, Ravi Kumar, Pasin Manurangsi, Mariana Raykova, and Phillipp Schoppmann. Distributed, private, sparse histograms in the two-server model. In CCS, pages 307-321, 2022. URL: https://doi.org/10.1145/3548606.3559383.
  5. Clément L Canonne, Gautam Kamath, and Thomas Steinke. The discrete gaussian for differential privacy. NeurIPS, 33:15676-15688, 2020. Google Scholar
  6. Ricardo Silva Carvalho, Ke Wang, and Lovedeep Gondara. Incorporating item frequency for differentially private set union. In AAAI, 2022. URL: https://api.semanticscholar.org/CorpusID:250289456.
  7. Justin Y. Chen, Vincent Cohen-Addad, Alessandro Epasto, and Morteza Zadimoghaddam. Scalable private partition selection via adaptive weighting. In ICML, 2025. URL: https://openreview.net/forum?id=NEfIaE4BI5.
  8. Damien Desfontaines, James Voss, Bryant Gipson, and Chinmoy Mandayam. Differentially private partition selection. PoPETS, 2022(1):339-352, November 2021. URL: https://doi.org/10.2478/popets-2022-0017.
  9. Google Differential Privacy Team. Privacy loss distributions, 2026. URL: https://github.com/google/differential-privacy/blob/main/.
  10. Vadym Doroshenko, Badih Ghazi, Pritish Kamath, Ravi Kumar, and Pasin Manurangsi. Connect the dots: Tighter discrete approximations of privacy loss distributions. PoPETS, 2022(4):552-570, 2022. URL: https://doi.org/10.56553/POPETS-2022-0122.
  11. Cynthia Dwork, Frank McSherry, Kobbi Nissim, and Adam Smith. Calibrating noise to sensitivity in private data analysis. In TCC, pages 265-284, 2006. URL: https://doi.org/10.1007/11681878_14.
  12. Badih Ghazi, Pritish Kamath, Ravi Kumar, and Pasin Manurangsi. Faster privacy accounting via evolving discretization. In ICML, pages 7470-7483, 2022. URL: https://proceedings.mlr.press/v162/ghazi22a.html.
  13. Atefeh Gilani, Juan Felipe Gomez, Shahab Asoodeh, Flavio P Calmon, Oliver Kosut, and Lalitha Sankar. Optimal additive noise mechanisms for differential privacy. arXiv preprint, 2025. URL: https://arxiv.org/abs/2504.14730.
  14. Sivakanth Gopi, Pankaj Gulhane, Janardhan Kulkarni, Judy Hanwen Shen, Milad Shokouhi, and Sergey Yekhanin. Differentially private set union. In ICML, pages 3627-3636, 2020. URL: https://proceedings.mlr.press/v119/gopi20a.html.
  15. Sivakanth Gopi, Yin Tat Lee, and Lukas Wutschitz. Numerical composition of differential privacy. In NeurIPS, pages 11631-11642, 2021. URL: https://proceedings.neurips.cc/paper/2021/hash/6097d8f3714205740f30debe1166744e-Abstract.html.
  16. Aleksandra Korolova, Krishnaram Kenthapadi, Nina Mishra, and Alexandros Ntoulas. Releasing search queries and clicks privately. In WWW, pages 171-180, 2009. URL: https://doi.org/10.1145/1526709.1526733.
  17. Antti Koskela, Joonas Jälkö, and Antti Honkela. Computing tight differential privacy guarantees using FFT. In AISTATS, pages 2560-2569, 2020. URL: http://proceedings.mlr.press/v108/koskela20b.html.
  18. Andrew L. Maas, Raymond E. Daly, Peter T. Pham, Dan Huang, Andrew Y. Ng, and Christopher Potts. Learning word vectors for sentiment analysis. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 142-150, Portland, Oregon, USA, June 2011. Association for Computational Linguistics. URL: http://www.aclweb.org/anthology/P11-1015.
  19. Julian McAuley and Jure Leskovec. Hidden factors and hidden topics: understanding rating dimensions with review text. Proceedings of the 7th ACM conference on Recommender systems, 2013. URL: https://api.semanticscholar.org/CorpusID:6440341.
  20. Sebastian Meiser and Esfandiar Mohammadi. Tight on budget? Tight bounds for r-fold approximate differential privacy. In CCS, pages 247-264, 2018. URL: https://doi.org/10.1145/3243734.3243765.
  21. Ilya Mironov. Rényi differential privacy. In CSF, pages 263-275, 2017. URL: https://doi.org/10.1109/CSF.2017.11.
  22. Nicolas Papernot and Thomas Steinke. Hyperparameter tuning with renyi differential privacy. In ICLR, 2022. URL: https://openreview.net/forum?id=-70L8lpp9DF.
  23. Alfréd Rényi. On measures of entropy and information. In Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Contributions to the Theory of Statistics, 1961. Google Scholar
  24. David M. Sommer, Sebastian Meiser, and Esfandiar Mohammadi. Privacy loss classes: The central limit theorem in differential privacy. PoPETS, 2019(2):245-269, 2019. URL: https://doi.org/10.2478/POPETS-2019-0029.
  25. Marika Swanberg, Damien Desfontaines, and Samuel Haney. Dp-sips: A simpler, more scalable mechanism for differentially private partition selection. PoPETS, 2023. Google Scholar
  26. Tim Van Erven and Peter Harremos. Rényi divergence and kullback-leibler divergence. IEEE Transactions on Information Theory, 60(7):3797-3820, 2014. URL: https://doi.org/10.1109/TIT.2014.2320500.
  27. Mark Wijkhuizen. Simple/normal wikipedia abstracts v1. URL: https://www.kaggle.com/datasets/markwijkhuizen/simplenormal-wikipedia-abstracts-v1.
  28. Royce J Wilson, Celia Yuxin Zhang, William Lam, Damien Desfontaines, Daniel Simmons-Marengo, and Bryant Gipson. Differentially private sql with bounded user contribution. PoPETS, 2:230-250, 2020. URL: https://doi.org/10.2478/POPETS-2020-0025.
  29. Xiang Zhang, Junbo Zhao, and Yann LeCun. Character-level convolutional networks for text classification. In C. Cortes, N. Lawrence, D. Lee, M. Sugiyama, and R. Garnett, editors, Advances in Neural Information Processing Systems, volume 28. Curran Associates, Inc., 2015. URL: https://proceedings.neurips.cc/paper_files/paper/2015/file/250cf8b51c773f3f8dc8b4be867a9a02-Paper.pdf.
  30. Yuqing Zhu, Jinshuo Dong, and Yu-Xiang Wang. Optimal accounting of differential privacy via characteristic function. In AISTATS, pages 4782-4817, 2022. URL: https://proceedings.mlr.press/v151/zhu22c.html.
Any Issues?
X

Feedback on the Current Page

CAPTCHA

Thanks for your feedback!

Feedback submitted to Dagstuhl Publishing

Could not send message

Please try again later or send an E-mail
© 2023-2026 Schloss Dagstuhl – LZI GmbH Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact