Our Data, Ourselves: Privacy Via Distributed Noise Generation
The DOI matches its Crossref record; 29 of the 31 references it cites were found in a registry or an index.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“Differential privacy (DP) [26] is an alternative theoretical model for protecting the privacy of individual data, which has been widely applied to many areas, not only traditional algorithms, e.g., boosting [27], principal component analysis [15], support vector machine [86], but also deep learning research [1, 70]. It ensures that the addition or removal does not substantially affect the outcome”
From Federated Learning for Healthcare Informatics (2021) · cites this paper as [26]
In the citing paper
“The simplest way to create a DP-mechanism is by adding Gaussian noise to the outcomes of a deterministic function with bounded sensitivity 21 .”
From Federated Learning and Differential Privacy for Medical Image Analysis (2021) · cites this paper as [21]
Citations
Cites
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
Self-citationThe Byzantine generals problem
Lamport, L., Shostak, R., & Pease, M. (1982) · ACM Transactions on Programming Languages and Systems · doi:10.1145/357172.357176 · Cited by 6,023 · cited 1×
“We will refer to several well-known primitive building blocks for constructing distributed protocols: Byzantine Agreement [20], Distributed Coin Flipping [22], Verifiable Secret Sharing (VSS) [8], Non-Malleable VSS, and Secure Function Evaluation (SFE) [18].”
Goldreich, O., Micali, S., & Wigderson, A. (1987) · Proceedings of the nineteenth annual ACM conference on Theory of computing - STOC '87 · doi:10.1145/28395.28420 · Cited by 3,590
Protocols for secure computations
Yao, A. C. (1982) · 23rd Annual Symposium on Foundations of Computer Science (sfcs 1982) · doi:10.1109/sfcs.1982.38 · Cited by 3,306
Privacy-preserving data mining
Agrawal, R., & Srikant, R. (2000) · ACM SIGMOD Record · doi:10.1145/335191.335438 · Cited by 2,961
Protocols for secure computations
Yao, A. C.-C. (1982) · doi:10.1109/sfcs.1982.88 · Cited by 2,661
Goldreich, O. (2001) · Cambridge University Press eBooks · doi:10.1017/cbo9780511546891 · Cited by 2,181
Goldreich, O. (2004) · Cambridge University Press eBooks · doi:10.1017/cbo9780511721656 · Cited by 1,805
Foundations of Cryptography: Volume 2, Basic Applications
Goldreich, O. (2004) · Cited by 1,592
On the design and quantification of privacy preserving data mining algorithms
Agrawal, D., & Aggarwal, C. C. (2001) · Proceedings of the twentieth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems · doi:10.1145/375551.375602 · Cited by 1,006 · cited 1×
“accurate statistics about a population while preserving the privacy of individuals [1, 2, 15, 11, 14, 5, 6, 4, 13].”
Cited by
Items 601–700 of 1,008
Differentially Private Kalman Filtering With Signal Aggregation
Degue, K. H., & Ny, J. L. (2022) · IEEE Transactions on Automatic Control · doi:10.1109/tac.2022.3230735 · Cited by 24
LDS-FL: Loss Differential Strategy Based Federated Learning for Privacy Preserving
Wang, T., Yang, Q., Zhu, K., Wang, J., Su, C., & Sato, K. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3322328 · Cited by 30
“It mainly adds some noise to the private dataset to make a small difference with that before [34], and then train the model on the dataset after encryption.”
BackgroundXie, C., Long, Y., Chen, P., Li, Q., Koyejo, S., & Li, B. (2023) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3576915.3623193 · Cited by 41
“Comparedtotest-timecerti � cationsagainstevasionattacks, training-time certi � cations against poisoning attacks have been less explored due to the notably di � erent threat models and the complexity of analyzing model training dynamics, even in a centralized setting.”
BackgroundMethodInfluentialSelective privacy-preserving framework for large language models fine-tuning
Wang, T., Zhai, L., Yang, T., Luo, Z., & Liu, S. (2024) · Information Sciences · doi:10.1016/j.ins.2024.121000 · Cited by 31
Learning from Data with Heterogeneous Noise using SGD.
Song, S., Chaudhuri, K., & Sarwate, A. D. (2015) · PubMed · Cited by 24
Distributed Hypothesis Testing with Privacy Constraints
Gilani, A., Amor, S. B., Salehkalaibar, S., & Tan, V. Y. F. (2019) · Entropy · doi:10.3390/e21050478 · Cited by 32
“Furthermore, a generalized notion— ( ǫ, δ ) -differential privacy [21]—provides a unified mathematical framework for data privacy.”
BackgroundDifferentially Private Publication of Vertically Partitioned Data
Tang, P., Cheng, X., Su, S., Chen, R., & Shao, H. (2019) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2019.2905237 · Cited by 24
“The schemes in [38] and [39] cannot provide sufficient privacy protection when a party is an attacker, while the scheme in [40] is highly inefficient for practical use.”
BackgroundDeriving an Optimal Noise Adding Mechanism for Privacy-Preserving Machine Learning
Kumar, M., Rossbory, M., Moser, B., & Freudenthaler, B. (2019) · Communications in computer and information science · doi:10.1007/978-3-030-27684-3_15 · Cited by 24
Differentially-Private Multi-Party Sketching for Large-Scale Statistics
Choi, S. G., Dachman-Soled, D., Kulkarni, M., & Yerukhimovich, A. (2020) · Proceedings on Privacy Enhancing Technologies · doi:10.2478/popets-2020-0047 · Cited by 24
Generating private synthetic databases for untrusted system evaluation
Lu, W., Miklau, G., & Gupta, V. (2014) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde.2014.6816689 · Cited by 24
“…single-table database, which adds Gaussian noise calibrated to the L 2 sensitivity (noted as ∆ ) to achieve ( (cid:15), δ ) -differential privacy [9], can be defined as: Definition 5.1 (Gaussian Mechanism): Assume Q contains d queries, the following randomized algorithm G provides ( (cid:15), δ )…”
BackgroundMethodInfluentialPrivate Identity Testing for High-Dimensional DistributionsPreprint
Canonne, C. L., Kamath, G., McMillan, A., Ullman, J., & Zakynthinou, L. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.11947 · Cited by 24
Fast Dimension Independent Private AdaGrad on Publicly Estimated SubspacesPreprint
Kairouz, P., Ribero, M., Rush, K., & Thakurta, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.06570 · Cited by 26
Privacy Amplification by Decentralization
Cyffers, E., & Bellet, A. (2020) · arXiv · doi:10.48550/arxiv.2012.05326 · Cited by 23
“A popular approach is to resort to cryptographic primitives to securely aggregate user contributions [14, 31, 7, 10, 21].”
BackgroundMethodBuilding a Secure and Privacy-Preserving Smart Grid
Birman, K., Jelasity, M., Kleinberg, R., & Tremel, E. (2015) · ACM SIGOPS Operating Systems Review · doi:10.1145/2723872.2723891 · Cited by 23
Privacy-preserving collaborative fuzzy clustering
Lyu, L., Bezdek, J. C., Law, Y. W., He, X., & Palaniswami, M. S. (2018) · Data & Knowledge Engineering · doi:10.1016/j.datak.2018.05.002 · Cited by 23
Equivalent mechanism: Releasing location data with errors through differential privacy
Wang, T., Zheng, Z., & Elhoseny, M. (2019) · Future Generation Computer Systems · doi:10.1016/j.future.2018.11.047 · Cited by 23
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 44
Add noise to remove noise: Local differential privacy for feature selection
Alishahi, M., Moghtadaiee, V., & Navidan, H. (2022) · Computers & Security · doi:10.1016/j.cose.2022.102934 · Cited by 23
HT-Fed-GAN: Federated Generative Model for Decentralized Tabular Data Synthesis
Duan, S., Liu, C., Han, P., Jin, X., Zhang, X., He, T., Pan, H., & Xiang, X. (2022) · Entropy · doi:10.3390/e25010088 · Cited by 30
“Differential Privacy Differential privacy [22,26] is a mathematical algorithm that indicates the level of privacy protection of individuals in a dataset.”
BackgroundTrading Utility for Privacy in Shared Spectrum Access Systems
Clark, M. A., & Psounis, K. (2017) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2017.2778260 · Cited by 23
“statistical database methods, including relaxed versions of differential privacy [9], offer some mechanisms to treat this”
MethodDong, T., Bu, X., & Hu, W. (2020) · Journal of the Franklin Institute · doi:10.1016/j.jfranklin.2019.12.027 · Cited by 23
Ghazi, B., Kumar, R., Manurangsi, P., & Pagh, R. (2021) · IT University Of Copenhagen (IT University of Copenhagen) · doi:10.48550/arxiv.2106.04247 · Cited by 23
“1: procedure ANALYZERD 2: U ← number of messages received 3: return U − E[D] Privacy.”
BackgroundInfluentialOnline Privacy-Preserving Data-Driven Network Anomaly Detection
Kurt, M. N., Yılmaz, Y., Wang, X., & Mosterman, P. J. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2022.3142302 · Cited by 23
“In [25], distributed implementation of privacy-preserving databases through distributed noise generation has been studied where each node generates a share of overall random noise.”
BackgroundDP-LSSGD: A Stochastic Optimization Method to Lift the Utility in Privacy-Preserving ERMPreprint
Wang, B., Gu, Q., Boedihardjo, M. T., Barekat, F., & Osher, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.12056 · Cited by 39
“These include the basic and strong composition theorems and their refinements [12, 14, 23], the moments accountant [1], etc.”
BackgroundMethodKnowledge Cross-Distillation for Membership Privacy
Chourasia, R., Enkhtaivan, B., Ito, K., Mori, J., Teranishi, I., & Tsuchida, H. (2022) · Proceedings on Privacy Enhancing Technologies · doi:10.2478/popets-2022-0050 · Cited by 23
“Known Defenses: MIAs can be mitigated using one known method, differential privacy [7, 8], which is a technique for guaranteeing worst-case privacy by adding noise to the learning objective or model outputs.”
MethodGroup Privacy for Personalized Federated Learning
Galli, F., Biswas, S., Jung, K., Cucinotta, T., & Palamidessi, C. (2023) · International Conference on Information Systems Security and Privacy (ICISSP) · doi:10.5220/0011885000003405 · Cited by 28
“One of the most successful approaches to address this issue in a rather robust way is along the lines of Differential Privacy (DP) [18, 19], which mathematically guarantees that a query output for a dataset does not change significantly regardless of whether a specific personal record is contained or…”
BackgroundInfluentialConcurrent Composition Theorems for Differential Privacy
Vadhan, S., & Zhang, W. (2023) · ACM Symposium on Theory of Computing (STOC) · doi:10.1145/3564246.3585241 · Cited by 22
“The basic composition theorem [4] states that the privacy parameters add up linearly when composing private mechanisms.”
BackgroundA Theory to Instruct Differentially-Private Learning via Clipping Bias Reduction
Xiao, H., Xiang, Z., Wang, D., & Devadas, S. (2023) · IEEE Symposium on Security and Privacy (S&P) · doi:10.1109/sp46215.2023.10179409 · Cited by 57
“Accordingly, the aggregation outputs are perturbed following Gaussian or Laplace mechanisms [8], [12].”
BackgroundMethodInfluentialSlaugFL: Efficient Edge Federated Learning With Selective GAN-Based Data Augmentation
Liu, J., Zhao, Z., Luo, X., Li, P., Min, G., & Li, H. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3397585 · Cited by 26
“For example, before uploading the local class prototype, we can add noise to it with the differential privacy technique [39] or directly encrypt it using cryptographic techniques [40].”
MethodImproving Privacy-Preserving Vertical Federated Learning by Efficient Communication with ADMM
Xie, C., Chen, P., Li, Q., Nourian, A., Zhang, C., & Li, B. (2024) · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) · doi:10.1109/satml59370.2024.00029 · Cited by 27
“In this work, we focus on the standard privacy notion (ε, δ)-DP [11, 10, 12].”
BackgroundA Comprehensive Location-Privacy-Awareness Task Selection Mechanism in Mobile Crowd-Sensing
Yan, K., Luo, G., Zheng, X., Tian, L., & Sai, A. M. V. V. (2019) · IEEE Access · doi:10.1109/access.2019.2921274 · Cited by 22
“It originated from the context of database [35]–[37].”
BackgroundDifferentially private data publishing via cross-moment microaggregation
Parra‐Arnau, J., Domingo‐Ferrer, J., & Soria-Comas, J. (2019) · Information Fusion · doi:10.1016/j.inffus.2019.06.011 · Cited by 22
Compressive Privacy for a Linear Dynamical System
Song, Y., Wang, C. X., & Tay, W. P. (2019) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2930366 · Cited by 23
“By adding white Gaussian noise ν k according to the Gaussian mechanism [59], Cz k is differentially private.”
BackgroundPrivEdge: From Local to Distributed Private Training and Prediction
Shamsabadi, A. S., Gascon, A., Haddadi, H., & Cavallaro, A. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.2988132 · Cited by 22
“Private training and prediction for MLaaS applications can be addressed by applying cryptography tools such as Fully Homomorphic Encryption (FHE) [31] and Multi-Party Computation (MPC) [32], as well as information theory and differential privacy [33].”
BackgroundMethodA Compression Perspective on Secrecy Measures
Shkel, Y. Y., & Poor, H. V. (2021) · IEEE Journal on Selected Areas in Information Theory · doi:10.1109/jsait.2021.3055692 · Cited by 24
Frequency estimation under multiparty differential privacy: one-shot and streaming
Huang, Z., Qiu, Y., Yi, K., & Cormode, G. (2022) · Proceedings of the VLDB Endowment · doi:10.14778/3547305.3547312 · Cited by 22
“It is also possible to consider computationally bounded adversaries [9, 24], but we will not explore that direction in this paper.”
BackgroundPrivate Adaptive Gradient Methods for Convex OptimizationPreprint
Asi, H., Duchi, J. C., Fallah, A., Javidbakht, O., & Talwar, K. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.13756 · Cited by 30
“We use the standard definitions of differential privacy (Dwork et al., 2006b;a): Definition 2.1.”
BackgroundMethodInfluentialTowards a Privacy-Aware Quantified Self Data Management Framework
Thuraisingham, B. M., Kantarcıoğlu, M., Bertino, E., Bakdash, J. Z., & Fernández, M. (2018) · ACM Symposium on Access Control Models and Technologies (SACMAT) · doi:10.1145/3205977.3205997 · Cited by 21
“As a part of this project, we will extend noise addition techniques combined with distributed cryptographic protocols (e.g., [20]) to produce differentially-private data mining results and leverage cloud infrastructure to scale to millions of individuals by delegating some work to cloud-based…”
MethodHow to Use Heuristics for Differential Privacy
Neel, S., Roth, A., & Wu, Z. S. (2019) · Annual Symposium on Foundations of Computer Science · doi:10.1109/focs.2019.00014 · Cited by 30
Concurrent Composition of Differential Privacy
Vadhan, S., & Wang, T. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-90453-1_20 · Cited by 41
Analysis of Application Examples of Differential Privacy in Deep Learning
Shen, Z., & Zhong, T. (2021) · Computational Intelligence and Neuroscience · doi:10.1155/2021/4244040 · Cited by 22
“In Definition 1, ε is the privacy budget [28], which represents the privacy protection level provided by the al-gorithm K .”
Protecting Sensitive Data in the Information Age: State of the Art and Future Prospects
Stach, C., Gritti, C., Bräcker, J., Behringer, M. H., & Mitschang, B. (2022) · Future Internet · doi:10.3390/fi14110302 · Cited by 25
“In this way, not only each discrete data point can be distorted but also, in the case of particularly strong noise, the progression of the time series itself [132].”
BackgroundGuaranteeing Differential Privacy in Distributed Fusion Estimation
Yan, X., Chen, B., Zhang, Y., & Yu, L. J. (2022) · IEEE Transactions on Aerospace and Electronic Systems · doi:10.1109/taes.2022.3219799 · Cited by 21
“Asabasicmechanism,noiseswithcertain distributionsareconsideredtoaffecttheprobabilityof publicly released estimates (PREs) after statistical query in differential privacy [14], [15].”
BackgroundAdaDpFed: A Differentially Private Federated Learning Algorithm With Adaptive Noise on Non-IID Data
Zhao, Z., Sun, Y., Bashir, A. K., & Lin, Z. (2023) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2023.3320122 · Cited by 21
“Dwork et al. [11] shows that ((cid:2), δ) -differential privacy is a relaxation of the plain (cid:2) -differential privacy with the broken probability δ .”
BackgroundBreaking the Trilemma of Privacy, Utility, and Efficiency via Controllable Machine Unlearning
Liu, Z., Dou, G., Chien, E., Zhang, C., Tian, Y., & Zhu, Z. (2024) · International World Wide Web Conference (WWW) · doi:10.1145/3589334.3645669 · Cited by 39
Samplable Anonymous Aggregation for Private Federated Data Analysis
Talwar, K., Wang, S., McMillan, A., Feldman, V., Bansal, P., Basile, B., Cahill, Á., Chan, Y. S., Chatzidakis, M., Chen, J., et al. (2024) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3658644.3690224 · Cited by 33
“We discuss additional related work in Section 6 and conclude with some open directions in Section 7.”
InfluentialA privacy-preserving task recommendation framework for mobile crowdsourcing
Gong, Y., Guo, Y., & Fang, Y. M. (2014) · Globecom · doi:10.1109/glocom.2014.7036871 · Cited by 21
“The focus of this paper is the task recommendation part, where tasks are selected and pushed to a user based on his context.”
BackgroundConvex Optimization for Linear Query Processing under Approximate Differential Privacy
Yuan, G., Yang, Y., Zhang, Z., & Hao, Z. (2016) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/2939672.2939818 · Cited by 21
“ly useful when the exa ct defi- nition is overly strict for an application, leading to poor result utility. One basic mechanism for enforcing approximate differential privacy is the Gaussian mechanism [4], which injects Gaussian noise to the query results calibrated to the ℓ2 sensitivity of the queries. For any two nei…”
BackgroundInherit Differential Privacy in Distributed Setting: Multiparty Randomized Function Computation
Wu, G., He, Y., Wu, J., & Xia, X. (2016) · IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom) · doi:10.1109/trustcom.2016.0157 · Cited by 21
“Although Protocol 3 in our paper is similar with the one in [23] to achieve Gaussian mechanism, our work focuses mainly on the fundamental theory and fundamental tools to compute randomized functions (and differentially private functions) in the distributed setting.”
MethodPrivacy-preserving obfuscation for distributed power systems
Mak, T. W. K., Fioretto, F., & Hentenryck, P. V. (2020) · Electric Power Systems Research · doi:10.1016/j.epsr.2020.106718 · Cited by 21
Investigating the Impact of Pre-trained Word Embeddings on Memorization in Neural Networks
Thomas, A., Adelani, D. I., Davody, A., Mogadala, A., & Klakow, D. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58323-1_30 · Cited by 43
εpsolute: Efficiently Querying Databases While Providing Differential Privacy
Bogatov, D., Kellaris, G., Kollios, G., Kobbi Nissim, B., & O’Neill, A. (2021) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3460120.3484786 · Cited by 21
Imtiaz, H., Mohammadi, J. A., Silva, R. F., Baker, B. T., Plis, S., Sarwate, A. D., & Calhoun, V. D. (2021) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2021.3126546 · Cited by 24
A Differentially Private Framework for Deep Learning With Convexified Loss Functions
Lü, Z., Asghar, H. J., Kâafar, M. A., Webb, D., & Dickinson, P. (2022) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3169911 · Cited by 28
“Table 4 shows the detailed configurations, where RDP, zCDP, AC and NC represent Rényi DP [28], zero-Concentrated DP [29], DP with advanced composition [21] and DP with naı̈ve composition [30], respectively, |X| is the size of the training set.”
BackgroundDifferential Privacy: A Cryptographic Approach to Private Data Analysis
Dwork, C. (2014) · Cambridge University Press eBooks · doi:10.1017/cbo9781107590205.018 · Cited by 25
Self-citationA New Class of Private Chi-Square TestsPreprint
Kifer, D., & Rogers, R. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.07662 · Cited by 22
“, to achieve concentrated differential privacy [5, 3] or (approximate) differential privacy [6]).”
BackgroundMethodInfluentialFast and Private Submodular and $k$-Submodular Functions Maximization with Matroid ConstraintsPreprint
Rafiey, A., & Yoshida, Y. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.15744 · Cited by 58
“Definition 2.4 (Dwork et al. (2006)) .”
InfluentialRobustness Threats of Differential PrivacyPreprint
Tursynbek, N., Petiushko, A., & Oseledets, I. V. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.07828 · Cited by 24
“The privacy parameters E and δ have the following interpretation: E is a privacy budget (maximal privacy loss) of an algorithm, and δ (generally, δ < 1/|D |), introduced in updated version [14], is the probability of possible violation of E-differential privacy.”
BackgroundPrivacy and Data-Based Research
Heffetz, O., & Ligett, K. (2013) · National Bureau of Economic Research · doi:10.3386/w19433 · Cited by 21
Liu, F. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1606.08052 · Cited by 21
“There are also softer versions of DP that guarantee privacy in a “weak” sense, including the ( (cid:15), δ )-approximate DP (aDP) (Dwork et al., 2006a), the ( (cid:15), δ )-probabilistic DP (pDP) (Machanavajjhala et al., 2008), the ( (cid:15), δ )-random DP (rDP) (Hall et al., 2012), and the (…”
BackgroundInfluentialPre-trained Perceptual Features Improve Differentially Private Image Generation
Harder, F., Asadabadi, M. J., Sutherland, D. J., & Park, M. (2022) · Transactions on machine learning research · doi:10.48550/arxiv.2205.12900 · Cited by 24
“∆μ is often called the global sensitivity (Dwork et al., 2006), which is the maximum difference in L2-norm given two neighbouring D and D′, ||μ(D)− μ(D)||2.”
BackgroundRAG Security and Privacy: Formalizing the Threat Model and Attack Surface
Arzanipour, A., Behnia, R., Ebrahimi, R., & Dutta, K. (2025) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw69685.2025.00165 · Cited by 20
“Definition 1 (Differential Privacy [25]): A randomized mechanism M : D → R is said to satisfy ( ε, δ ) -differential privacy if, for any pair of adjacent datasets X, X ′ ∈ D , and for every measurable subset S ⊆ R , the following holds: Here, adjacency between datasets is defined by the relation X…”
Private classification with limited labeled data
Liu, X., Li, Q., & Li, T. (2017) · Knowledge-Based Systems · doi:10.1016/j.knosys.2017.07.006 · Cited by 20
Differentially Private Normalizing Flows for Synthetic Tabular Data Generation
Lee, J. W., Kim, M., Jeong, Y., & Ro, Y. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i7.20697 · Cited by 32
“Definition 1 (( ϵ, δ )-DP (Dwork et al. 2006b,a)) .”
An Operational Approach to Information Leakage via Generalized Gain Functions
Kurri, G. R., Sankar, L., & Kosut, O. (2023) · IEEE Transactions on Information Theory · doi:10.1109/tit.2023.3341148 · Cited by 26
“Approximate differential privacy [46] and Rényi differential privacy [47] are proposed as relaxations of differential privacy to allow data releases with higher utility.”
BackgroundAccuracy-Privacy Tradeoffs for Two-Party Differentially Private Protocols
Goyal, V., Mironov, I., Pandey, O., & Sahai, A. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-40041-4_17 · Cited by 20
“We now recall the definition of ε-differential privacy [10] and (ε, δ)-differential privacy [11].”
BackgroundSelf-citationTowards Differential Privacy for Symbolic Systems
Jones, A. M., Leahy, K., & Hale, M. (2019) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc.2019.8814723 · Cited by 20
The optimal upper bound of the number of queries for Laplace mechanism under differential privacy
Li, X., Li, H., Zhu, H., & Huang, M. (2019) · Information Sciences · doi:10.1016/j.ins.2019.07.001 · Cited by 20
A Novel Cloud-Assisted Secure Deep Feature Classification Framework for Cancer Histopathology Images
Kumar, A., Singh, S. K., Lakshmanan, K. S., Saxena, S., & Shrivastava, S. (2021) · ACM Transactions on Internet Technology · doi:10.1145/3424221 · Cited by 20
An accurate, scalable and verifiable protocol for federated differentially private averaging
Sabater, C., Bellet, A., & Ramon, J. (2022) · Machine Learning · doi:10.1007/s10994-022-06267-9 · Cited by 43
“per party MSE Verif Risks Central DP [36] O(1) O(1/n2) No Trusted curator Local DP [55] O(1) O(1/n) No Verifiable secret sharing [35] O(n) O(1/n2) Yes Secure agg.”
BackgroundMethodInfluentialAn Antifolk Theorem for Large Repeated Games
Pai, M. M., Roth, A., & Ullman, J. (2016) · ACM Transactions on Economics and Computation · doi:10.1145/2976734 · Cited by 20
“8 Formally, we have the following theorem due to Dwork et al. [2010].”
BackgroundMethodInfluentialA study of privacy and fairness in sensitive data analysis
Barak, B., & Hardt, M. (2011) · Cited by 20
DP-CGAN: Differentially Private Synthetic Data and Label GenerationPreprint
Torkzadehmahani, R., Kairouz, P., & Paten, B. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.09700 · Cited by 38
“Differential privacy [10, 11] is a mathematical framework to express the level of privacy preservation of individuals in a statistical databases.”
BackgroundMethodInfluentialA comprehensive survey on privacy-preserving technologies for Smart Grids
Bibi, H., Abolhasan, M., Lipman, J., Abdollahi, M., & Ni, W. (2025) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2025.110371 · Cited by 25
Verified Foundations for Differential Privacy
de Medeiros, M., Naveed, M., Lepoint, T., Kahsai, T., Ravitch, T., Zetzsche, S., Joshi, A., Tassarotti, J., Albarghouthi, A., & Tristan, J.-B. (2025) · Proceedings of the ACM on Programming Languages · doi:10.1145/3729294 · Cited by 19
Privacy-Preserving Collaborative Data Collection and Analysis with Many Missing Values
Sei, Y., Andrew, J., Okumura, H., & Ohsuga, A. (2022) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2022.3174887 · Cited by 25
“Multiparty computation is a technology that requires long computation time but can process data confidentially [87].”
BackgroundDefense against membership inference attack in graph neural networks through graph perturbation
Wang, K., Wu, J., Zhu, T., Ren, W., & Hong, Y. (2022) · International Journal of Information Security · doi:10.1007/s10207-022-00646-y · Cited by 19
“Compare to differential privacy technique [34–36], which also make use of random noise, but from the same data distribution (i.e., using the same set of parameters), we propose to observe the trade-off between graph privacy and usability caused by the noise from different data distributions.”
BackgroundRobust and privacy-preserving collaborative training: a comprehensive survey
Yang, F., Zhang, X., Guo, S., Chen, D., Gan, Y., Xiang, T., & Liu, Y. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10797-0 · Cited by 23
Privacy-Preserving Machine Learning in Cloud–Edge–End Collaborative Environments
Yang, W., Wang, H., Li, Z., Niu, Z., Wu, L., Wei, X., Su, Y., & Susilo, W. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3461410 · Cited by 25
“Specifically, Xie et al. [30] is used to initialize privacy budgets and perform subsequent optimizations, Abadi et al. [28] is employed to determine variances based on privacy budgets, and Dwork et al. [41] is utilized for secure generation of random Gaussian noise in a distributed manner.”
Chowdhery, A., Levorato, M., Burago, I., & Baidya, S. (2017) · Springer eBooks · doi:10.1007/978-3-319-57639-8_6 · Cited by 19
Differentially-Private Two-Party Egocentric Betweenness Centrality
Roohi, L., Rubinstein, B. I. P., & Teague, V. (2019) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom.2019.8737405 · Cited by 19
Privacy-preserving Crowd-guided AI Decision-making in Ethical Dilemmas
Wang, T., Zhao, J., Yu, H., Liu, J., Yang, X., Ren, X., & Shi, S. (2019) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3357384.3357954 · Cited by 20
“Definition 1 (ε-Differential Privacy (DP) [10, 12]).”
BackgroundMethodDifferentially Private Robust ADMM for Distributed Machine Learning
Ding, J., Zhang, X., Chen, M., Xue, K., Zhang, C., & Pan, M. (2019) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata47090.2019.9005716 · Cited by 23
“2) Differential Privacy: Differential privacy (DP) as an important privacy paradigm is first presented in [12], which provides robust protection against a wide range of attacks by injecting random noise to perturb the released statistical results obtained from sensitive datasets.”
BackgroundThe limits of pan privacy and shuffle privacy for learning and estimation
Cheu, A., & Ullman, J. (2021) · ACM Symposium on Theory of Computing (STOC) · doi:10.1145/3406325.3450995 · Cited by 27
“Although there are more practical protocols implementing certain differentially private algorithms ([22] et seq.”
BackgroundDifferential privacy protection over deep learning: An investigation of its impacted factors
Lin, Y., Bao, L.-Y., Li, Z.-M., Si, S.-Z., & Chu, C. (2020) · Computers & Security · doi:10.1016/j.cose.2020.102061 · Cited by 21
Instance-optimality in differential privacy via approximate inverse sensitivity mechanisms
Asi, H., & Duchi, J. C. (2020) · neural information processing systems · Cited by 87
Cost-Friendly Differential Privacy of Smart Meters Using Energy Storage and Harvesting Devices
Hossain, M. B., Natgunanathan, I., Xiang, Y., & Zhang, Y. (2021) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2021.3081170 · Cited by 19
“ADP is a relaxation of standard DP [20] that can be defined as follows [22]: For all datasets, C 1 and C 2 differing by only one element, a query S gives ð (cid:2); d Þ DP if and only if for all Y 2 Range ð S Þ where the output of S is denoted by Range ð S Þ .”
BackgroundPrivate and Online Learnability Are Equivalent
Alon, N., Bun, M., Livni, R., Malliaris, M., & Moran, S. (2022) · Journal of the ACM · doi:10.1145/3526074 · Cited by 63
“[31, 32] If p,q are (ε,δ )-indistinguishable then for all k ∈ N, p and q are (kε,kδ )indistinguishable, where p ,q are the k-fold products of p,q (i.”
BackgroundMeasuring Linkability of Protected Biometric Templates Using Maximal Leakage
Shahreza, H. O., Shkel, Y. Y., & Marcel, S. (2023) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2023.3266170 · Cited by 23
“The most prominent notions of privacy are ϵ -differential privacy and (ϵ, δ) - differential privacy which were developed for the database release problem [20], [21], [22].”
BackgroundBarrientos, A. F., Williams, A. R., Snoke, J., & Bowen, C. M. (2023) · Journal of the American Statistical Association · doi:10.1080/01621459.2023.2270795 · Cited by 19
“Several relaxations of (cid:2) -DP have been developed in order to inject less noise into the output, such as ((cid:2) , δ) -DP (Dwork et al. 2006), probabilistic DP (Machanavajjhala et al. 2008), concentrated DP (Dwork and Rothblum 2016), Rényi DP (Mironov 2017), and zero-concentrated DP (Bun and…”
BackgroundMethodInfluentialGenerating highly accurate prediction hypotheses through collaborative ensemble learning
Arsov, N., Pavlovski, M., Basnarkov, L., & Kocarev, L. (2017) · Scientific Reports · doi:10.1038/srep44649 · Cited by 19
“Recently, algorithmic stability has been also connected to differential privacy 40 , (robust and perfect) generalization 6 , adaptive data analysis 41 , adaptive learning and compression schemes 42 .”
Sensitivity Analysis for Non-Interactive Differential Privacy: Bounds and Efficient Algorithms
İnan, A., Gürsoy, M. E., & Saygın, Y. (2017) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2017.2734664 · Cited by 19
“The most notable relaxation is ð"; dÞ-DP [22], where Definition 2 would instead be written as: Pr1⁄2AðDÞ 2 S e Pr1⁄2AðD0Þ 2 S þ d.”
Using Deep Learning to Generate Relational HoneyData
Abay, N. C., Akçora, C. G., Zhou, Y., Kantarcıoğlu, M., & Thuraisingham, B. M. (2019) · Springer eBooks · doi:10.1007/978-3-030-02110-8_1 · Cited by 22
Privacy-preserving set-based estimation using partially homomorphic encryption
Alanwar, A., Gaßmann, V., He, X., Said, H., Sandberg, H., Johansson, K. H., & Althoff, M. (2023) · European Journal of Control · doi:10.1016/j.ejcon.2023.100786 · Cited by 19
“To overcome the addition of excessive noise, a combination of homomorphic encryption with distributed noise has been proposed in [39], where each estimator generated its share of the aggregated noise required for di ff erential privacy [40] and sent encrypted and obfuscated data to the aggregator.”
BackgroundMachine Learning, Reasoning, and Intelligence in Daily Life: Directions and Challenges
Horvitz, E. (2006) · Cited by 19
Model-based Differentially Private Data SynthesisPreprint
Liu, F. (2016) · arXiv (Cornell University) · Cited by 22
Graph-based Clustering under Differential PrivacyPreprint
Pinot, R., Morvan, A., Yger, F., Gouy‐Pailler, C., & Atif, J. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1803.03831 · Cited by 19
“Definition 2.4 (reformulation Dwork et al. (2006b)) .”
InfluentialElement Level Differential Privacy: The Right Granularity of PrivacyPreprint
Asi, H., Duchi, J. C., & Javidbakht, O. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.04042 · Cited by 22
“We now turn to the problem of estimating item frequencies—histogram estimation—one of the original motivations for differential privacy [Dwork et al. 2006b, Ex.”
InfluentialA Survey on Federated Analytics: Taxonomy, Enabling Techniques, Applications and Open Issues
Wang, Z., Ji, H., Zhu, Y., Wang, D., & Han, Z. (2025) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2025.3558755 · Cited by 24
“The idea of DDP is inspired by [137], and the primary formal DDP designs is proposed in [138], [139], which are much earlier than the introduction of the term “DDP”.”
A Framework for Adversarial Streaming Via Differential Privacy and Difference Estimators
Attias, I., Cohen, E., Shechner, M., & Stemmer, U. (2024) · Algorithmica · doi:10.1007/s00453-024-01259-8 · Cited by 18
“% That is, from now on, all TDE’s at levels i < j are estimating differences from future time steps to the current time step.”
Background
Metadata
- Authors
- Dwork, C., Kenthapadi, K., McSherry, F., Mironov, I., & Naor, M.
- Affiliations
- Microsoft (United States) (US)
- Stanford University (US)
- Weizmann Institute of Science (IL)
- Venue
- Lecture Notes in Computer Science
- Publisher
- Springer Berlin Heidelberg
- Published
- 2006
- Pages
- 486-503
- ISSN
- 0302-9743 (ISSN-L)
- 1611-3349
- OpenAlex
- W1557833142
- Cited by
- 1,835
- References
- 31
- Citations by year
- 2026: 65 · 2025: 138 · 2024: 184 · 2023: 179 · 2022: 175 · 2021: 188 · 2020: 216 · 2019: 123 · 2018: 80 · 2017: 81 · 2016: 61 · 2015: 69 · 2014: 78 · 2013: 70 · 2012: 52
- FWCI
- 7.51
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Lecture notes in computer science (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › Cryptography and Data Security
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Information Systems › Cloud Data Security Solutions
- Keywords
- malicious security, verifiable secret sharing, Gaussian noise, exponential distribution, random bit generation, distributed protocols
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.