Filters








17,856 Hits in 6.3 sec

Testing Differential Privacy with Dual Interpreters [article]

Hengchu Zhang, Edo Roth, Andreas Haeberlen, Benjamin C. Pierce, Aaron Roth
<span title="2020-10-08">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Applying differential privacy at scale requires convenient ways to check that programs computing with sensitive data appropriately preserve privacy.  ...  We propose here a fully automated framework for testing differential privacy, adapting a well-known "pointwise" technique from informal proofs of differential privacy.  ...  TESTING DIFFERENTIAL PRIVACY To show how DPCheck tests differential privacy, we first review how proofs of differential privacy are constructed (5.1), with ReportNoisyMax as an example (5.2).  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2010.04126v1">arXiv:2010.04126v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/gefyg52w6ffonamm7frwetwtau">fatcat:gefyg52w6ffonamm7frwetwtau</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201017160720/https://arxiv.org/pdf/2010.04126v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/af/88/af88d623c2e3cbb1eefd6bf61a155bce27545e2a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2010.04126v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Testing Differential Privacy with Dual Interpreters [article]

Hengchu Zhang, Edo Roth, Andreas Haeberlen, Benjamin Pierce, Aaron Roth
<span title="2020-09-17">2020</span> <i title="Zenodo"> Zenodo </i> &nbsp;
This archive contains all files and documentation submitted to OOPSLA 2020 AEC for the paper "Testing Differential Privacy with Dual Interpreters".  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.5281/zenodo.4034723">doi:10.5281/zenodo.4034723</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/v3yjqpfyznedhjhipajqg2j33q">fatcat:v3yjqpfyznedhjhipajqg2j33q</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200920044651/https://zenodo.org/record/4034724/files/Overview.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/6c/14/6c14eaf73fadbf6ff427b2d8d5700835c966e20c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.5281/zenodo.4034723"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> zenodo.org </button> </a>

Sharp Composition Bounds for Gaussian Differential Privacy via Edgeworth Expansion [article]

Qinqing Zheng, Jinshuo Dong, Qi Long, Weijie J. Su
<span title="2020-03-25">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
This raises a fundamental question in differential privacy regarding how the overall privacy bound degrades under composition.  ...  To address this question, we introduce a family of analytical and sharp privacy bounds under composition using the Edgeworth expansion in the framework of the recently proposed f-differential privacy.  ...  This privacy definition leverages the hypothesis testing interpretation of differential privacy, and characterizes the privacy guarantee using the trade-off between type I and type II errors given by the  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.04493v2">arXiv:2003.04493v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/46fecmsfdjhmronvjfbh7vc7ti">fatcat:46fecmsfdjhmronvjfbh7vc7ti</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200902132614/https://arxiv.org/pdf/2003.04493v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/5c/0a/5c0ad85dda6824eadc693f606daf127791d227b7.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2003.04493v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Preserving privacy between features in distributed estimation

Christina Heinze-Deml, Brian McWilliams, Nicolai Meinshausen
<span title="">2018</span> <i title="Wiley"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/mgn2wo4tmzc7vm7yg6mhkh7qg4" style="color: black;">Stat</a> </i> &nbsp;
We work with the notion of (ϵ,δ)-distributed differential privacy which extends single-party differential privacy to the distributed, vertically-partitioned case.  ...  In this setting few approaches exist for private data sharing for the purposes of statistical estimation and the classical setup of differential privacy with a "trusted curator" preparing the data does  ...  When S = {1, . . . , p}, S-differential privacy reduces to ( , δ)-differential privacy.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1002/sta4.189">doi:10.1002/sta4.189</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/nqss46i5prdgvjr2sm5qz2yrza">fatcat:nqss46i5prdgvjr2sm5qz2yrza</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20191021044416/https://arxiv.org/pdf/1703.00403v1.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/6f/b3/6fb3a9a4ae62a25ca699c08840696eef14855b47.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1002/sta4.189"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> wiley.com </button> </a>

Optimal Noise Adding Mechanisms for Approximate Differential Privacy

Quan Geng, Pramod Viswanath
<span title="">2016</span> <i title="Institute of Electrical and Electronics Engineers (IEEE)"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/niovmjummbcwdg4qshgzykkpfu" style="color: black;">IEEE Transactions on Information Theory</a> </i> &nbsp;
functions, and show that the ( , δ)-differential privacy is a framework not much more general than the ( , 0)-differential privacy and (0, δ)-differential privacy in the context of 1 and 2 cost functions  ...  We conclude that in ( , δ)-differential privacy, the optimal noise magnitude and the noise power are (min((1/ ), (1/δ))) and (min((1/ 2 ), (1/δ 2 ))), respectively, in the high privacy regime.  ...  Operational Meaning of ( , δ)-Differential Privacy in the Context of Hypothesis Testing We first give an operational interpretation of differential privacy in the context of hypothesis testing.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tit.2015.2504972">doi:10.1109/tit.2015.2504972</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/c43fnek6jzexvhnynwgbjprzbm">fatcat:c43fnek6jzexvhnynwgbjprzbm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20171026052030/http://pramodv.ece.illinois.edu:80/pubs/GV16_2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/a0/8c/a08c1c3203126f24d07a800d0dfec3cf9a360827.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tit.2015.2504972"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>

Differentially Private Continual Learning [article]

Sebastian Farquhar, Yarin Gal
<span title="2019-02-18">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We present a differentially private continual learning framework based on variational inference.  ...  Catastrophic forgetting can be a significant problem for institutions that must delete historic data for privacy reasons. For example, hospitals might not be able to retain patient data permanently.  ...  It is this generative model which we then train with attention to differential privacy.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1902.06497v1">arXiv:1902.06497v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/6h2qhhkk65fkflk7qmqolgvauy">fatcat:6h2qhhkk65fkflk7qmqolgvauy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20191015002559/https://arxiv.org/pdf/1902.06497v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/e9/cb/e9cb98d3e88496026db8d025798c33ce8fd39ac6.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1902.06497v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Differentially Private Collaborative Intrusion Detection Systems For VANETs [article]

Tao Zhang, Quanyan Zhu
<span title="2020-05-02">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We use the differential privacy to capture the privacy notation of the PML-CIDS and propose a method of dual variable perturbation to provide dynamic differential privacy.  ...  One fundamental barrier to collaborative learning is the privacy concern as nodes exchange data among them.  ...  Then, Alg-2 can be interpreted as Z dual v ( f v , τ|D v , ε) := Z v ( f v |D v ) + C 1 n v ε f v .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.00703v1">arXiv:2005.00703v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/tnd6yzhj5vhjxorudvcum3o7gq">fatcat:tnd6yzhj5vhjxorudvcum3o7gq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200506100932/https://arxiv.org/pdf/2005.00703v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2005.00703v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Optimal Noise Adding Mechanisms for Approximate Differential Privacy [article]

Quan Geng, Pramod Viswanath
<span title="2013-12-19">2013</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
general than the (ϵ,0)-differential privacy and (0,δ)-differential privacy in the context of ℓ^1 and ℓ^2 cost functions, i.e., minimum expected noise magnitude and noise power.  ...  We conclude that in (ϵ,δ)-differential privacy, the optimal noise magnitude and noise power are Θ((1/ϵ,1/δ)) and Θ((1/ϵ^2,1/δ^2)), respectively, in the high privacy regime.  ...  Operational Meaning of ( , δ)-differential privacy in the Context of Hypothesis Testing As shown by [27] , one can interpret the differential privacy constraint (3) in the context of hypothesis testing  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1305.1330v3">arXiv:1305.1330v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hj2j6uqor5gcrcshss7g54s3jm">fatcat:hj2j6uqor5gcrcshss7g54s3jm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200909032604/https://arxiv.org/pdf/1305.1330v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c6/5e/c65e8bbdb69db41c06eebd4b2d37796c78b4a476.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1305.1330v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

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

Peter Kairouz, Sewoong Oh, Pramod Viswanath
<span title="2014-10-07">2014</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
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  ...  The optimality is very general: it holds for all types of functions, heterogeneous privacy conditions on the parties, all types of cost metrics, and both average and worst-case (over the inputs) measures  ...  It follows from the next theorem on the operational interpretation of differential privacy.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1407.1546v2">arXiv:1407.1546v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/t62lufc32zdifgrcmpqsp4kg6e">fatcat:t62lufc32zdifgrcmpqsp4kg6e</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200826042254/https://arxiv.org/pdf/1407.1546v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/f2/93/f293066a9c50ad5c793021c615502c3de4488c10.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1407.1546v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Investigating the Personalization–Privacy Paradox in Internet of Things (IoT) Based on Dual-Factor Theory: Moderating Effects of Type of IoT Service and User Value

Ae-Ri Lee
<span title="2021-09-26">2021</span> <i title="MDPI AG"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/oglosmy3gbhuzobyjit4qalakq" style="color: black;">Sustainability</a> </i> &nbsp;
It aims to analyze the impact of the dual factor—personalization and privacy concerns related to IoT services—on the intention to use IoT.  ...  Further, the model includes four-dimensional motivated innovativeness and previous privacy-invasion experience as key antecedents of the dual factor.  ...  The extent of common method bias was tested with Harman's one-factor test [84] , which assesses whether a single factor accounts for greater than 50% of the variance.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/su131910679">doi:10.3390/su131910679</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3ynulse24bcx5atmxa2zmyx4ym">fatcat:3ynulse24bcx5atmxa2zmyx4ym</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210926114229/https://mdpi-res.com/d_attachment/sustainability/sustainability-13-10679/article_deploy/sustainability-13-10679.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/13/11/131192e4a804c3c91cc32ce742220644aaa22cf5.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.3390/su131910679"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> mdpi.com </button> </a>

Differentially Private and Fair Deep Learning: A Lagrangian Dual Approach [article]

Cuong Tran, Ferdinando Fioretto, Pascal Van Hentenryck
<span title="2020-09-26">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
The method relies on the notion of differential privacy and the use of Lagrangian duality to design neural networks that can accommodate fairness constraints while guaranteeing the privacy of sensitive  ...  The paper analyses the tension between accuracy, privacy, and fairness and the experimental evaluation illustrates the benefits of the proposed model on several prediction tasks.  ...  The idea is to render the computations of the primal and dual update steps differentially private with respect to the sensitive attributes.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2009.12562v1">arXiv:2009.12562v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/objnqxrywves5pgqxbfk25fgci">fatcat:objnqxrywves5pgqxbfk25fgci</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200930011103/https://arxiv.org/pdf/2009.12562v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2009.12562v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Dynamic Privacy For Distributed Machine Learning Over Network [article]

Tao Zhang, Quanyan Zhu
<span title="2016-03-09">2016</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
dynamic differential privacy.  ...  The two mechanisms lead to algorithms that can provide privacy guarantees under mild conditions of the convexity and differentiability of the loss function and the regularizer.  ...  Then, Alg-2 can be interpreted as minimizing Z dual p ( f p ,t|D p , ε pi (t)) with initial condition as f p (0) = f p (t) and λ p (0) = λ p (t) for all p ∈ P.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1601.03466v3">arXiv:1601.03466v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/2w2aesxmsrbolg2lch6kbg3jlq">fatcat:2w2aesxmsrbolg2lch6kbg3jlq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200825161134/https://arxiv.org/pdf/1601.03466v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c0/04/c004a3365ef1d3f45ef82027cb263d11c19a1dc6.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1601.03466v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Research on wireless distributed financial risk data stream mining based on dual privacy protection

Yuhao Zhao
<span title="2020-11-27">2020</span> <i title="Springer Science and Business Media LLC"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/a5gjzjadkzbirdua4xdqipotp4" style="color: black;">EURASIP Journal on Wireless Communications and Networking</a> </i> &nbsp;
This article is mainly aimed at the current privacy data leakage in financial data mining, combined with existing data mining technology to study data mining and privacy protection.  ...  First, a data mining model for dual privacy protection is defined, which can better meet the characteristics of distributed data streams while achieving privacy protection effects.  ...  This is consistent with the interpretation of the New Palgrave Dictionary of Monetary and Financial Affairs.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1186/s13638-020-01842-x">doi:10.1186/s13638-020-01842-x</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/2po2gbhl4zfr3lh5i2bnosnj5m">fatcat:2po2gbhl4zfr3lh5i2bnosnj5m</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201129223740/https://jwcn-eurasipjournals.springeropen.com/track/pdf/10.1186/s13638-020-01842-x.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/d9/8d/d98d60d46d2c11bd64353e5d50cd4e84d5aa71b7.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1186/s13638-020-01842-x"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> springer.com </button> </a>

Optimized Query Forgery for Private Information Retrieval

David Rebollo-Monedero, Jordi Forne
<span title="">2010</span> <i title="Institute of Electrical and Electronics Engineers (IEEE)"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/niovmjummbcwdg4qshgzykkpfu" style="color: black;">IEEE Transactions on Information Theory</a> </i> &nbsp;
We carefully justify and interpret our privacy criterion from diverse perspectives.  ...  Our formulation poses a mathematically tractable problem that bears substantial resemblance with rate-distortion theory.  ...  Parra-Arnau, with the Department of Telematics Engineering at the Technical University of Catalonia, the editor of this journal, and the anonymous reviewers for their careful reading and helpful comments  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tit.2010.2054471">doi:10.1109/tit.2010.2054471</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/opjln3kb3zavlj7w7loh6vnaou">fatcat:opjln3kb3zavlj7w7loh6vnaou</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20170817205658/http://upcommons.upc.edu/bitstream/handle/2117/10267/OptimizedQueryForgery.pdf?sequence=1" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/ed/50/ed50b60f523a452d847ff1c6b738574e4fa33ae4.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1109/tit.2010.2054471"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> ieee.com </button> </a>

Deep Learning with Gaussian Differential Privacy [article]

Zhiqi Bu and Jinshuo Dong and Qi Long and Weijie J. Su
<span title="2020-07-22">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we consider a recently proposed privacy definition termed f-differential privacy [18] for a refined privacy analysis of training neural networks.  ...  An increasingly important line of work therefore has sought to train neural networks subject to privacy constraints that are specified by differential privacy or its divergence-based relaxations.  ...  This new privacy definition faithfully retains the hypothesis testing interpretation of differential privacy and can losslessly reason about common primitives associated with differential privacy, including  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1911.11607v3">arXiv:1911.11607v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/nd27ura2efdxncz63owe7sc4g4">fatcat:nd27ura2efdxncz63owe7sc4g4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200725022925/https://arxiv.org/pdf/1911.11607v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1911.11607v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 17,856 results