Automatic Discovery of Privacy–Utility Pareto Fronts
Open accessAbstract
Differential privacy is a mathematical framework for privacy-preserving data analysis. Changing the hyperparameters of a differentially private algorithm allows one to trade off privacy and utility in a principled way. Quantifying this trade-off in advance is essential to decision-makers tasked with deciding how much privacy can be provided in a particular application while maintaining acceptable utility. Analytical utility guarantees offer a rigorous tool to reason about this tradeoff, but are generally only available for relatively simple problems. For more complex tasks, such as training neural networks under differential privacy, the utility achieved by a given algorithm can only be measured empirically. This paper presents a Bayesian optimization methodology for efficiently characterizing the privacy– utility trade-off of any differentially private algorithm using only empirical measurements of its utility. The versatility of our method is illustrated on a number of machine learning tasks involving multiple models, optimizers, and datasets.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Utilisation of DP-SGD unavoidably negatively affects the utility of resulting models, a privacy-utility trade-off.
“the authors of this and subsequent works noted that the utilisation of DP-SGD unavoidably negatively affects the utility of the resulting models, a well-known effect termed the privacy-utility trade-off13.”
From Medical imaging deep learning with differential privacy (2021) · cites this paper as [13]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · see citing works
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · see citing works
Calibrating Noise to Sensitivity in Private Data Analysis
Dwork, C., McSherry, F., Nissim, K., & Smith, A. (2006) · Lecture Notes in Computer Science · doi:10.1007/11681878_14 · Cited by 7,272 · see citing works
Practical Bayesian Optimization of Machine Learning Algorithms
Snoek, J., Larochelle, H., & Adams, R. P. (2012) · arXiv · doi:10.48550/arxiv.1206.2944 · Cited by 6,691 · see citing works
Deep Learning with Differential Privacy
Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016) · Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2976749.2978318 · Cited by 6,578 · see citing works
Dwork, C. (2006) · Lecture notes in computer science · doi:10.1007/11787006_1 · Cited by 5,945 · see citing works
Large-Scale Machine Learning with Stochastic Gradient Descent
Bottou, L. (2010) · Proceedings of COMPSTAT'2010 · doi:10.1007/978-3-7908-2604-3_16 · Cited by 5,795 · see citing works
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2014) · Foundations and Trends® in Theoretical Computer Science · doi:10.1561/0400000042 · Cited by 5,557 · see citing works
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2013) · doi:10.1561/9781601988195 · Cited by 3,856 · see citing works
Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning)
Rasmussen, C. E., & Williams, C. K. I. (2005) · The MIT Press eBooks · Cited by 1,829 · see citing works
MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
Chen, T., Li, M., Li, Y., Lin, M., Wang, N., Wang, M., Xiao, T., Xu, B., Zhang, C., & Zhang, Z. (2015) · arXiv · doi:10.48550/arxiv.1512.01274 · Cited by 1,819 · see citing works
Scaling up the accuracy of Naive-Bayes classifiers: a decision-tree hybrid
Kohavi, R. (1996) · Cited by 1,404 · see citing works
Hyperband: A Novel Bandit-Based Approach to Hyperparameter OptimizationPreprint
Li, L., Jamieson, K., DeSalvo, G., Rostamizadeh, A., & Talwalkar, A. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1603.06560 · Cited by 1,368 · see citing works
Mironov, I. (2017) · 2017 IEEE 30th Computer Security Foundations Symposium (CSF) · doi:10.1109/csf.2017.11 · Cited by 1,255 · see citing works
On Bayesian Methods for Seeking the Extremum
Mockus, J. (1974) · Optimization Techniques · doi:10.1007/3-540-07165-2_55 · Cited by 1,172 · see citing works
Hyperband: a novel bandit-based approach to hyperparameter optimization
Li, L., Jamieson, K., DeSalvo, G., Rostamizadeh, A., & Talwalkar, A. (2017) · Journal of Machine Learning Research · Cited by 930 · see citing works
Learning Differentially Private Recurrent Language Models
McMahan, H. B., Ramage, D., Talwar, K., & Zhang, L. (2017) · arXiv · doi:10.48550/arxiv.1710.06963 · Cited by 849 · see citing works
Kernels for Vector-Valued Functions: a Review
Álvarez, M. A., Rosasco, L., & Lawrence, N. D. (2012) · White Rose Research Online (University of Leeds, The University of Sheffield, University of York) · Cited by 587 · see citing works
Golovin, D., Solnik, B., Moitra, S., Kochanski, G., Karro, J. E., & Sculley, D. (2017) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3097983.3098043 · Cited by 585 · see citing works
Convergence rates of efficient global optimization algorithmsPreprint
Bull, A. D. (2011) · arXiv (Cornell University) · doi:10.5555/1953048.2078198 · Cited by 454 · see citing works
Fast Bayesian Optimization of Machine Learning Hyperparameters on Large DatasetsPreprint
Klein, A., Falkner, S., Bartels, S., Hennig, P., & Hutter, F. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.07079 · Cited by 424 · see citing works
Kernels for Vector-Valued Functions: A Review
Álvarez, M. A., Rosasco, L., & Lawrence, N. D. (2012) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000036 · Cited by 418 · see citing works
On Bayesian Methods for Seeking the Extremum
Mockus, A. (1975) · Springer eBooks · doi:10.1007/978-3-662-38527-2_55 · Cited by 410 · see citing works
On the complexity of differentially private data release
Dwork, C., Naor, M., Reingold, O., Rothblum, G. N., & Vadhan, S. (2009) · ACM Symposium on Theory of Computing (STOC) · doi:10.1145/1536414.1536467 · Cited by 363 · see citing works
Snelson, E., Rasmussen, C. E., Zoubin Ghahramani, A., Thrun, S., Saul, L. K., & Schölkopf, B. (2003) · Cited by 257 · see citing works
Couckuyt, I., Deschrijver, D., & Dhaene, T. (2013) · Journal of Global Optimization · doi:10.1007/s10898-013-0118-2 · Cited by 252 · see citing works
Subsampled Rényi Differential Privacy and Analytical Moments Accountant
Wang, Y.-X., Balle, B., & Kasiviswanathan, S. P. (2020) · Journal of Privacy and Confidentiality · doi:10.29012/jpc.723 · Cited by 225 · see citing works
Self-citationSubsampled Rényi Differential Privacy and Analytical Moments Accountant
Wang, Y.-X., Balle, B., & Kasiviswanathan, S. (2018) · arXiv · doi:10.48550/arxiv.1808.00087 · Cited by 224
Bolt-on Differential Privacy for Scalable Stochastic Gradient Descent-based Analytics
Wu, X., Li, F., Kumar, A., Chaudhuri, K., Jha, S., & Naughton, J. F. (2017) · ACM SIGMOD International Conference on Management of Data (SIGMOD) · doi:10.1145/3035918.3064047 · Cited by 182 · see citing works
An economic analysis of privacy protection and statistical accuracy as social choices
Abowd, J. M., & Schmutte, I. M. (2019) · American Economic Review · doi:10.1257/aer.20170627 · Cited by 162 · see citing works
Privacy Amplification by Subsampling: Tight Analyses via Couplings and DivergencesPreprint
Balle, B., Barthe, G., & Gaboardi, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1807.01647 · Cited by 154 · see citing works
Self-citationUnderstanding the sparse vector technique for differential privacy
Lyu, M., Su, D., & Li, N. (2017) · Proceedings of the VLDB Endowment · doi:10.14778/3055330.3055331 · Cited by 129 · see citing works
Efficient Hyperparameter Optimization and Infinitely Many Armed BanditsPreprint
Rostamizadeh, A., Talwalkar, A., DeSalvo, G., Jamieson, K., & Li, L. (2017) · arXiv (Cornell University) · Cited by 127 · see citing works
A General Approach to Adding Differential Privacy to Iterative Training Procedures
McMahan, H. B., Andrew, G., Erlingsson, U., Chien, S., Mironov, I., Papernot, N., & Kairouz, P. (2018) · arXiv · doi:10.48550/arxiv.1812.06210 · Cited by 126 · see citing works
The Secret Sharer: Measuring Unintended Neural Network Memorization & Extracting SecretsPreprint
Carlini, N., Liu, C., Kos, J., Erlingsson, Ú., & Song, D. X. (2018) · arXiv (Cornell University) · Cited by 124 · see citing works
Privacy Amplification by Iteration
Feldman, V., Mironov, I., Talwar, K., & Thakurta, A. (2018) · Annual Symposium on Foundations of Computer Science · doi:10.1109/focs.2018.00056 · Cited by 118 · see citing works
Issues Encountered Deploying Differential Privacy
Garfinkel, S. L., Abowd, J. M., & Powazek, S. (2018) · Proceedings of the 2018 Workshop on Privacy in the Electronic Society · doi:10.1145/3267323.3268949 · Cited by 112 · see citing works
Balle, B., & Wang, Y.-X. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.06530 · Cited by 77 · see citing works
Self-citationPrivate selection from private candidates
Liu, J., & Talwar, K. (2019) · ACM SIGACT Symposium on Theory of Computing (STOC) · doi:10.1145/3313276.3316377 · Cited by 66 · see citing works
GPflowOpt: A Bayesian Optimization Library using TensorFlowPreprint
Knudde, N., van der Herten, J., Dhaene, T., & Couckuyt, I. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.03845 · Cited by 53 · see citing works
Dispersion for Data-Driven Algorithm Design, Online Learning, and Private Optimization
Balcan, M.-F., Dick, T., & Vitercik, E. (2018) · Annual Symposium on Foundations of Computer Science · doi:10.1109/focs.2018.00064 · Cited by 51 · see citing works
Accuracy First: Selecting a Differential Privacy Level for Accuracy-Constrained ERMPreprint
Ligett, K., Neel, S., Roth, A., Waggoner, B., & Wu, Z. S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1705.10829 · Cited by 47 · see citing works
Kotsogiannis, I., Machanavajjhala, A., Hay, M., & Miklau, G. (2017) · ACM SIGMOD International Conference on Management of Data (SIGMOD) · doi:10.1145/3035918.3035945 · Cited by 36 · see citing works
Targeting solutions in Bayesian multi-objective optimization: sequential and batch versions
Gaudrie, D., Riche, R. L., Picheny, V., Enaux, B., & Herbert, V. (2019) · Annals of Mathematics and Artificial Intelligence · doi:10.1007/s10472-019-09644-8 · Cited by 34 · see citing works
Ge, C., He, X., Ilyas, I. F., & Machanavajjhala, A. (2019) · ACM SIGMOD International Conference on Management of Data (SIGMOD) · doi:10.1145/3299869.3300092 · Cited by 33 · see citing works
Rényi Differential Privacy Mechanisms for Posterior SamplingPreprint
Geumlek, J., Song, S., & Chaudhuri, K. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1710.00892 · Cited by 28 · see citing works
Differentially Private Bayesian OptimizationPreprint
Kusner, M. J., Gardner, J. R., Garnett, R., & Weinberger, K. Q. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1501.04080 · Cited by 27 · see citing works
The Large Margin Mechanism for Differentially Private MaximizationPreprint
Chaudhuri, K., Hsu, D., & Song, S. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1409.2177 · Cited by 26 · see citing works
Differentially private regression with Gaussian processes
Smith, M. T., Álvarez, M. A., Zwiessele, M., & Lawrence, N. D. (2018) · White Rose Research Online (University of Leeds, The University of Sheffield, University of York) · Cited by 23 · see citing works
A Batched Scalable Multi-Objective Bayesian Optimization AlgorithmPreprint
Lin, X., Zhen, H.-L., Li, Z., Zhang, Q., & Kwong, S. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.01323 · Cited by 10 · see citing works
Private Selection from Private CandidatesPreprint
Liu, J., & Talwar, K. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.07971 · Cited by 3 · see citing works
APEx: Accuracy-Aware Differentially Private Data ExplorationPreprint
Ge, C., He, X., Ilyas, I. F., & Machanavajjhala, A. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1712.10266 · Cited by 3 · see citing works
Balancing exploration and exploitation in multiobjective batch bayesian optimization
Wang, H., Xu, H., Yuan, Y., Sun, X., & Deng, J. (2019) · Genetic and Evolutionary Computation Conference Companion (GECCO Companion) · doi:10.1145/3319619.3321962 · Cited by 3 · see citing works
Flagged first, then by citations
1 of the 54 references couldn’t be matched to a known paper.
Research works printed but not linked: 6
- [24][24] Rodolphe Jenatton, Cedric Archambeau, Javier González, and Matthias Seeger. Bayesian optimization with treestructured dependencies. In Doina Precup and Yee Whye Teh, editors, Proceedings of the 34th International Conference on Machine Learning , volume 70 of Proceedings of Machine Learning Research , pages 1655-1664, International Convention Centre, Sydney, Australia, 06-11 Aug 2017. PMLR. -
- [29][29] Ios Kotsogiannis, Ashwin Machanavajjhala, Michael Hay, and Gerome Miklau. Pythia: Data dependent differentially private algorithm selection. In Proceedings of the 2017 ACM International Conference on Management of Data , pages 1323-1337. ACM, 2017. -
- [39][39] Ilya Mironov. Rényi differential privacy. In Computer Security Foundations Symposium (CSF), 2017 IEEE 30th , pages 263-275. IEEE, 2017. -
- [41][41] Kobbi Nissim, Thomas Steinke, Alexandra Wood, Micah Altman, Aaron Bembenek, Mark Bun, Marco Gaboardi, David O'Brien, and Salil Vadhan. Differential privacy: A primer for a non-technical audience (preliminary version). Vanderbilt Journal of Entertainment and Technology Law , 2018. -
- [46][46] Koen Lennart van der Veen. A practical approach to differential private learning. Master's thesis, University of Amsterdam, The Netherlands, 2018. -
- [50][50] Marcela Zuluaga, Andreas Krause, and Markus Püschel. ε -pal: an active learning approach to the multi-objective optimization problem. The Journal of Machine Learning Research , 17(1):3619-3650, 2016.
Cited by
Sorted by citations
Challenges in deploying machine learning: A survey of case studies
Paleyes, A., Urma, R.-G., & Lawrence, N. D. (2023) · ACM Computing Surveys · doi:10.1145/3533378 · Cited by 638 · see citing works
Self-citationMedical imaging deep learning with differential privacy
Ziller, A., Usynin, D., Braren, R., Makowski, M., Rueckert, D., & Kaissis, G. (2021) · Scientific Reports · doi:10.1038/s41598-021-93030-0 · Cited by 172 · see citing works
The Role of Adaptive Optimizers for Honest Private Hyperparameter Selection
Mohapatra, S., Sasy, S., He, X. Q., Kamath, G., & Thakkar, O. D. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i7.20749 · Cited by 15 · see citing works
Efficient Privacy-Utility Optimization for Differentially Private Deep Learning
Rafika, B., Fethallah, H., & Belloum, A. S. Z. (2025) · International journal of electrical and computer engineering systems · doi:10.32985/ijeces.16.5.3 · Cited by 5 · see citing works
Partial sensitivity analysis in differential privacyPreprint
Mueller, T. T., Ziller, A., Usynin, D., Knolle, M. A., Jungmann, F., Rueckert, D., & Kaissis, G. A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.10582
- Authors
- Avent, B., González, J., Diethe, T., Paleyes, A., & Balle, B.
- Affiliations
- University of Southern California (US)
- Microsoft (United States) (US)
- Amazon (United States) (US)
- University of Cambridge (GB)
- Google DeepMind (United Kingdom) (GB)
- Venue
- Proceedings on Privacy Enhancing Technologies
- Published
- 2020-08-17
- Volume
- 2020
- Issue
- 4
- Pages
- 5-23
- ISSN
- 2299-0984 (ISSN-L)
- OpenAlex
- W2946348439
- Cited by
- 3
- References
- 54
- Citations by year
- 2025: 1 · 2022: 1 · 2021: 1
- FWCI
- 0.20
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Proceedings on Privacy Enhancing Technologies (publishedVersion, cc-by-nc-nd)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques
- Social Sciences › Sociology and Political Science › Privacy, Security, and Data Protection
- Keywords
- differential privacy, privacy-utility trade-off, Pareto front, Bayesian optimization, hyperparameter optimization, neural network training, privacy-preserving machine learning, multi-objective optimization
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
