A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf
.
Filters
Differential Privacy without Sensitivity
2016
Neural Information Processing Systems
The exponential mechanism is a general method to construct a randomized estimator that satisfies (ε, 0)-differential privacy. ...
In this paper, we focus on (ε, δ)-differential privacy of Gibbs posteriors with convex and Lipschitz loss functions. ...
Differential privacy without sensitivity In this section, we state our main results for (ε, δ)-differential privacy in the form of Claim 1. ...
dblp:conf/nips/MinamiASN16
fatcat:y34qgxebnvghtmb53voohh6ahe
A Gentle Introduction to Differential Privacy with Use Cases
2021
Zenodo
A gentle introduction to Differential Privacy as a small summary of how to understand it for non technical experts. ...
sources: • I want to know more about DP without getting in detail: • Why differential privacy is awesome -Damien Desfontaines • Understanding differential privacy and why it matters for digital rights ...
Differential Privacy • Differential Privacy (DP) is a property that gives privacy guarantees to user of a population. ...
doi:10.5281/zenodo.5101323
fatcat:s7hxj2bcizccxhvaxaiaxy6rpa
A Gentle Introduction to Differential Privacy with Use Cases
2021
Zenodo
A gentle introduction to Differential Privacy as a small summary of how to understand it for non technical experts. ...
sources: • I want to know more about DP without getting in detail: • Why differential privacy is awesome -Damien Desfontaines • Understanding differential privacy and why it matters for digital rights ...
Differential Privacy • Differential Privacy (DP) is a property that gives privacy guarantees to user of a population. ...
doi:10.5281/zenodo.5095316
fatcat:wddodaccqjfr3o5fa4m3vb6pse
Differential Privacy Technique for Privacy Preservation on Big Data
2019
International Journal for Research in Applied Science and Engineering Technology
In this article, we proposed a novel mechanism, called Adaptive Firefly Laplace Mechanism (AFLM), to preserve differential privacy on Big Data to protect sensitive information among analyst. ...
To overcome the drawbacks of existing privacy preservation methods such as both cryptographic techniques and data anonymization are analyzed with differential privacy method in this article.Privacy has ...
Figure 1: Differential Privacy Preservation Differential privacy method provides strong protection to sensitive data or personal data in big data processing. ...
doi:10.22214/ijraset.2019.6049
fatcat:nocjjhvhffgs3gs2h3lzxmts2y
Solo: A Lightweight Static Analysis for Differential Privacy
[article]
2021
arXiv
pre-print
All current approaches for statically enforcing differential privacy in higher order languages make use of either linear or relational refinement types. ...
We demonstrate such an embedding in Haskell, demonstrate its expressiveness on case studies, and prove that our type-based enforcement of differential privacy is sound. ...
This distinction allows the implementation of S 's privacy monad in Haskell, and additionally enables our approach to describe variants of differential privacy without linear group privacy (e.g. ( , )differential ...
arXiv:2105.01632v2
fatcat:2fd2pfx3ibaullw7z75a5otyqu
Not Just Cloud Privacy: Protecting Client Privacy in Teacher-Student Learning
[article]
2020
arXiv
pre-print
In this work, we re-design the privacy-preserving "teacher-student" model consisting of adopting both private arbitrary masking and local differential privacy, which protects the sensitive information ...
Ensuring the privacy of sensitive data used to train modern machine learning models is of paramount importance in many areas of practice. ...
It aims at providing provable privacy guarantee for each sensitive data sample, unlike general differential privacy is protecting the whole sensitive dataset [3] . ...
arXiv:1910.08038v2
fatcat:6xu7fcftjncgxleadbnkvhbvcu
On syntactic anonymity and differential privacy
2013
2013 IEEE 29th International Conference on Data Engineering Workshops (ICDEW)
Differential privacy has been promoted as the answer to privacy-preserving data mining. We discuss here issues involved and criticisms of both approaches, and conclude that both have their place. ...
We identify research directions that will enable greater access to data while improving privacy guarantees. ...
without violating differential privacy; the promise of differential privacy is (by design) not absolute secrecy. ...
doi:10.1109/icdew.2013.6547433
dblp:conf/icde/CliftonT13
fatcat:ydfypwstnfgpxe74yo37rdo6ja
Differentially Private Variational Autoencoders with Term-wise Gradient Aggregation
[article]
2020
arXiv
pre-print
This paper studies how to learn variational autoencoders with a variety of divergences under differential privacy constraints. ...
Using differentially private SGD (DP-SGD), which randomizes a stochastic gradient by injecting a dedicated noise designed according to the gradient's sensitivity, we can easily build a differentially private ...
to cover the increased sensitivity fail into differential privacy guarantee that we expected. ...
arXiv:2006.11204v1
fatcat:ars4wmrbnrcw3iyjehgrqy7cjq
Differential Privacy Preserving Genomic Data Releasing via Factor Graph
[chapter]
2017
Lecture Notes in Computer Science
privacy noise directly injected into lowdimensional local distributions 1
1
t
1
2
t
1
11
Injection of Differential Privacy Noise
16
n items, each
with sensitivity
3/m
r items, each ...
SNP position: BB, Bb, or bb.
9
Differential Privacy [DMNS, TCC 06]
9
A
10
Differential Privacy
10
To construct approximate distribution p * (X U |X K , F, A). m: # individuals in ...
doi:10.1007/978-3-319-59575-7_33
fatcat:vvo6d44rpbgl7k4b7yymjugrhu
On the Intrinsic Differential Privacy of Bagging
[article]
2020
arXiv
pre-print
Differentially private machine learning trains models while protecting privacy of the sensitive training data. ...
In particular, we prove that, for any base learner, Bagging with and without replacement respectively achieves (N· k ·lnn+1/n,1- (n-1/n)^N· k)-differential privacy and (lnn+1/n+1-N· k, N· k/n)-differential ...
using public non-sensitive data in a privacy-preserving way. ...
arXiv:2008.09845v1
fatcat:inxf6va5wvbd3l2qr57iwu2vxy
Differential privacy for eye tracking with temporal correlations
2021
PLoS ONE
Our results provide significantly high privacy without any essential loss in classification accuracies while hiding personal identifiers. ...
Privacy-preservation techniques such as differential privacy mechanisms have recently been applied to eye movement data obtained from such displays. ...
Privacy sensitivity classification tasks for MPIIPrivacEye are carried only without majority voting since privacy sensitivity of the scene can change at each time step and applying majority voting to such ...
doi:10.1371/journal.pone.0255979
pmid:34403454
pmcid:PMC8370645
fatcat:3d3tu465rzcinjs3lqqeciom3q
Controlling Privacy Loss in Survey Sampling (Working Paper)
[article]
2020
arXiv
pre-print
This intuition has been formalized in the differential privacy literature for simple random sampling: a differentially private mechanism run on a simple random subsample of a population provides higher ...
We find that not only do these schemes often not amplify privacy, but that they can result in privacy degradation. ...
there be a variant of differential privacy (with similar semantics) that does enjoy amplification under data-dependent sampling without replacement? ...
arXiv:2007.12674v1
fatcat:ghgf43dzdnbs3l3ip3cqc4jlya
Differential Privacy-enabled Federated Learning for Sensitive Health Data
[article]
2020
arXiv
pre-print
Second, it uses a differential privacy mechanism to further protect the model from potential privacy attacks. ...
sensitive data, resource constraints for transferring and integrating data from multiple sites, and risk of a single point of failure. ...
More specifically, we show that FL without differential privacy can provide model performance close to the hypothetical case where all data is centralized. ...
arXiv:1910.02578v3
fatcat:7okfkjznyvb6fdway4attytbxe
COVID-19 Imaging Data Privacy by Federated Learning Design: A Theoretical Framework
[article]
2020
arXiv
pre-print
In this paper, we introduce differential privacy by design (dPbD) framework and discuss its embedding into the federated machine learning system. ...
We discuss the evaluation of the proposed design of federated machine learning systems and discuss how differential privacy by design (dPbD) framework can enhance data privacy in federated learning systems ...
• How can differentially-private federated learning systems ensure the privacy of COVID-19 imaging data without adversely harming the accuracy? ...
arXiv:2010.06177v1
fatcat:asfm6ub2krc7bj3p2vyugqfdiu
Blockchain-Based Differential Privacy Cost Management System
[article]
2020
arXiv
pre-print
Privacy preservation is a big concern for various sectors. To protect individual user data, one emerging technology is differential privacy. ...
Blockchain will be able to keep track of all noisy responses generated with differential privacy algorithm and allow for certain queries to reuse old responses. ...
Differential privacy algorithm scales the noise generated with the sensitivity of the query function. ...
arXiv:2006.04693v1
fatcat:yalzkevmcvapnpbdcubudnvkxa
« Previous
Showing results 1 — 15 out of 96,840 results