Filters








176,949 Hits in 3.7 sec

Differentially-private Federated Neural Architecture Search [article]

Ishika Singh, Haoyi Zhou, Kunlin Yang, Meng Ding, Bill Lin, Pengtao Xie
2020 arXiv   pre-print
To further preserve privacy, we study differentially-private FNAS (DP-FNAS), which adds random noise to the gradients of architecture variables.  ...  To address this problem, we propose federated neural architecture search (FNAS), where different parties collectively search for a differentiable architecture by exchanging gradients of architecture variables  ...  Differential Privacy Rajkumar and Agarwal (Rajkumar and Agarwal, 2012) developed differentially-private machine learning algorithms in a distributed multi-party setting.  ... 
arXiv:2006.10559v2 fatcat:7xpjs4bc6rf5nj6ljle4mnzdpm

Inherit Differential Privacy in Distributed Setting: Multiparty Randomized Function Computation [article]

Genqiang Wu and Yeping He and Jingzheng Wu and Xianyao Xia
2016 arXiv   pre-print
How to achieve differential privacy in the distributed setting, where the dataset is distributed among the distrustful parties, is an important problem.  ...  A composition theorem of differentially private protocols is also presented.  ...  U (0, 1) in the two-party setting.  ... 
arXiv:1604.03001v1 fatcat:jg6piqufj5e2zendmvbqy5oase

Shrinkwrap: Differentially-Private Query Processing in Private Data Federations [article]

Johes Bater, Xi He, William Ehrich, Ashwin Machanavajjhala, Jennie Rogers
2018 arXiv   pre-print
A private data federation is a set of autonomous databases that share a unified query interface offering in-situ evaluation of SQL queries over the union of the sensitive data of its members.  ...  We introduce Shrinkwrap, a private data federation that offers data owners a differentially private view of the data held by others to improve their performance over oblivious query processing.  ...  M-Party Support In this work, we implement Shrinkwrap using two party secure computation protocols, meaning that we run our experiments over two data owners.  ... 
arXiv:1810.01816v1 fatcat:lg4a56xs7vfnhmvwpiox3qqym4

Shrinkwrap

Johes Bater, Xi He, William Ehrich, Ashwin Machanavajjhala, Jennie Rogers
2018 Proceedings of the VLDB Endowment  
A private data federation is a set of autonomous databases that share a unified query interface offering in-situ evaluation of SQL queries over the union of the sensitive data of its members.  ...  We introduce Shrinkwrap, a private data federation that offers data owners a differentially private view of the data held by others to improve their performance over oblivious query processing.  ...  M-Party Support In this work, we implement Shrinkwrap using two party secure computation protocols, meaning that we run our experiments over two data owners.  ... 
doi:10.14778/3291264.3291274 fatcat:32upgt2bdfb3bc2c2diswsste4

PrivMin: Differentially Private MinHash for Jaccard Similarity Computation [article]

Ziqi Yan, Jiqiang Liu, Gang Li, Zhen Han, Shuo Qiu
2017 arXiv   pre-print
Then for achieving strict differential privacy in MH-JSC, we propose the PrivMin algorithm, which consists of two private operations: 1) the Private MinHash Value Generation that works by introducing the  ...  To achieve this, we first define the Conditional $\epsilon$-DPSO, a relaxed differential privacy definition regarding set operations, and prove that the MinHash-based Jaccard Similarity Computation (MH-JSC  ...  Based on the above setting, the Differentially Private Set Operations ( -DPSO) is formally defined as Definition 6 ( -DPSO).  ... 
arXiv:1705.07258v1 fatcat:un7aw5etgzabbptexi5erhrppa

DP-Cryptography: Marrying Differential Privacy and Cryptography in Emerging Applications [article]

Sameer Wagh, Xi He, Ashwin Machanavajjhala, Prateek Mittal
2020 arXiv   pre-print
There are two popular models for deploying differential privacy - standard differential privacy (SDP), where a trusted server aggregates all the data and runs the DP mechanisms, and local differential  ...  Second, we synthesize work in an area we call "DP-Cryptography" - cryptographic primitives that are allowed to leak differentially private outputs.  ...  This protocol applies private setintersection cardinality technique to privately compute the noisy intersection set of the two datasets.  ... 
arXiv:2004.08887v1 fatcat:xkwpazhrd5gyndl2sbs3pquxni

Privacy-Preserving Multi-Party Contextual Bandits [article]

Awni Hannun, Brian Knott, Shubho Sengupta, Laurens van der Maaten
2020 arXiv   pre-print
This paper develops a privacy-preserving multi-party contextual bandit for this learning setting by combining secure multi-party computation with a differentially private mechanism based on epsilon-greedy  ...  This paper considers a learning setting in which multiple parties aim to train a contextual bandit together in a private way: the parties aim to maximize the total reward but do not want to share any of  ...  The AND operation between two private values, x y , is implemented akin to the private multiplication protocol using Beaver triples. ] ← θ z + [β xr ] − [θ r ] − [η xr ] (number of wraps in [x]).  ... 
arXiv:1910.05299v3 fatcat:4yt2qxezifgo3ofqylodcpa6cy

Secure Mining Of Association Rules In Distributed Datasets

Qilong Han, Dan Lu, Kejia Zhang, Hongtao Song, Haitao Zhang
2019 IEEE Access  
INDEX TERMS Data mining, distributed databases, association rule mining, differential privacy, privacy preserving.  ...  The existing solutions for distributed data(vertical partition and horizontal partition) have high complexity of encryption and incomplete definition of attributes of multiple parties.  ...  Furthermore, the solution in [10] only works for three parties or more, not for two parties.  ... 
doi:10.1109/access.2019.2948033 fatcat:7mesct5qzfbqzo7rod3ficr73y

Combining Differential Privacy and Secure Multiparty Computation

Martin Pettai, Peeter Laud
2015 Proceedings of the 31st Annual Computer Security Applications Conference on - ACSAC 2015  
We consider how to perform privacy-preserving analyses on private data from different data providers and containing personal information of many different individuals.  ...  We combine differential privacy and secret sharing in the same system to protect the privacy of both the data providers and the individuals.  ...  They use an expensive offline preprocessing phase, which in the online phase allows multiplying two integers with each party sending only two values to every other party (as opposed to five in Sharemind  ... 
doi:10.1145/2818000.2818027 dblp:conf/acsac/PettaiL15 fatcat:vwuyawwonzgurngwbqe7yeks2a

Privacy Preserving Record Linkage via grams Projections [article]

Luca Bonomi, Li Xiong, Rui Chen, Benjamin C. M. Fung
2012 arXiv   pre-print
In this paper, we study the problem of private record linkage via secure data transformations.  ...  In contrast to the existing techniques in this area, we propose a novel approach that provides strong privacy guarantees under the formal framework of differential privacy.  ...  These sets are sent to the third party C. 4.  ... 
arXiv:1208.2773v1 fatcat:ews7lbwdevhevd5yfi5n3z3ij4

A privacy framework

Jinfei Liu, Li Xiong, Jun Luo
2013 Proceedings of the Joint EDBT/ICDT 2013 Workshops on - EDBT '13  
privacy w.r.t. semi-honest behavior in the secure multi-party computation setting, and prove they are equivalent.  ...  With decades' development, there are three essential privacy principles in the literature on PPDP and SMC: Bayes-optimal privacy [24] in the PPDP setting, and differential privacy [6] in the PPDP setting  ...  Given an operation of private data (e.g. a data release, a statistical computation, or a multi-party securecomputation), intercepting the operation knowledge should give an adversary no or little (ϵ) new  ... 
doi:10.1145/2457317.2457340 dblp:conf/edbt/LiuXL13 fatcat:ah33u74vz5f2xa7nwk2f7rwvsq

Achieving Differential Privacy in Vertically Partitioned Multiparty Learning [article]

Depeng Xu, Shuhan Yuan, Xintao Wu
2019 arXiv   pre-print
Preserving differential privacy has been well studied under centralized setting.  ...  However, it's very challenging to preserve differential privacy under multiparty setting, especially for the vertically partitioned case.  ...  DPFW achieves (ε, δ)-differential privacy and has two versions, DPFW-C in the centralized setting and DPFW-K in the multiparty setting. We specify the number of parties K=2,4,8 in our comparison.  ... 
arXiv:1911.04587v1 fatcat:knjswozvvfggvlj2hbhvp6a4o4

Differentially Private Multi-party Computation: Optimality of Non-Interactive Randomized Response [article]

Peter Kairouz, Sewoong Oh, Pramod Viswanath
2014 arXiv   pre-print
We study the problem of interactive function computation by multiple parties possessing a single bit each in a differential privacy setting (i.e., there remains an uncertainty in any specific party's bit  ...  Each party is interested in computing a function, which could differ from party to party, and there could be a central observer interested in computing a separate function.  ...  If the protocol is λ i -differentially private for all i ∈ [k], then we say that the protocol is {λ i }-differentially private for all parties.  ... 
arXiv:1407.1546v2 fatcat:t62lufc32zdifgrcmpqsp4kg6e

Differential Privacy in Multi-Party Resource Sharing [article]

Utku Karaca, S. Ilker Birbil, Nursen Aydin, Gizem Mullaoglu
2021 arXiv   pre-print
In this setting, the concern for the parties is the privacy of their data and their optimal allocations. We propose a two-step approach to meet the privacy requirements of the parties.  ...  In the second step, we provide this guarantee with a differentially private algorithm at the expense of deviating slightly from the optimality.  ...  The set of constraints (2) guarantees that the total consumptions of the parties do not exceed the shared capacities, whereas the set of constraints (3) expresses the private constraints of party k ∈ K  ... 
arXiv:2110.10498v1 fatcat:qkrlf7ph6reqnleqahnuhq4gti

The Privacy-preserving Padding Problem: Non-negative Mechanisms for Conservative Answers with Differential Privacy [article]

Benjamin M. Case and James Honaker and Mahnush Movahedi
2021 arXiv   pre-print
We show how these mechanisms can be applied in a few select areas including making the cardinalities of set intersections and unions revealed in Private Set Intersection protocols differential private  ...  Differentially private noise mechanisms commonly use symmetric noise distributions.  ...  Application to Private Set Intersection Private Set Intersection considers the problem where two parties X and Y hold private sets D X , D Y ⊂ D and wish to compute some function on the intersection of  ... 
arXiv:2110.08177v1 fatcat:defdozivp5d5zisjgpgdntlxfa
« Previous Showing results 1 — 15 out of 176,949 results