2026-09-19T13:48:14Zhttps://dspace.mit.edu/server/oai/request

oai:dspace.mit.edu:1721.1/1241072021-07-05T14:03:20Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023

Srini Devadas. Xiao, Hanshen. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science 2020-03-09T18:53:48Z 2020-03-09T18:53:48Z 2019 2019 https://hdl.handle.net/1721.1/124107 1142812131 Thesis: S.M., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2019 Cataloged from PDF version of thesis. Includes bibliographical references (pages 79-83). Privacy concerns with sensitive data are receiving increasing attention. In this thesis, we study local differential privacy (LDP) in interactive decentralized optimization. Comparing to central differential privacy (DP), where a centralized curator maintains the dataset, LDP is a stronger notion yet with industrial adoption, which allows data of an individual to be privatized before sharing. Consequently, more challenges are encountered to build efficient statistical analyzer in LDP setting. Towards practical decentralized optimization in LDP, we extend LDP into a more comprehensive notion which provides both worst and average case privacy guarantees. Accordingly, two approaches to sharpen utility-privacy tradeoff are proposed for the worst and the average, respectively: First, cryptographically incorporated with merely linear secret sharing, we show the privacy guarantee can be improved by a factor of [square root of] N' where N' amongst all N agents are semi-honest. Second, we take Alternating Direction Method of Multipliers (ADMM), and decentralized (stochastic) gradient descent(D(S)GD) as two concrete examples to propose a framework of first-order based optimization with random local aggregators. We prove such local randomization lead to the same utility guarantee but amplify average LDP by a constant, empirically around 30%. Thorough experiments support our theory. by Hanshen Xiao. S.M. S.M. Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science 83 pages eng Massachusetts Institute of Technology MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission. http://dspace.mit.edu/handle/1721.1/7582 Electrical Engineering and Computer Science. Local differential privacy in decentralized optimization Thesis application/pdf 2020-03-09T18:53:47Z Publication Master EECS unknown unknown <Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="a1e829f1-7c36-4d39-9dc0-2537122953bd"> <Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type> <Language>eng</Language> <Title>Local differential privacy in decentralized optimization</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2019</PublicationDate> <Authors> <Author> <DisplayName>Xiao, Hanshen.</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <DisplayName>Massachusetts Institute of Technology</DisplayName> <OrgUnit /> </Publisher> </Publishers> <License>http://dspace.mit.edu/handle/1721.1/7582</License> <Keyword>Electrical Engineering and Computer Science.</Keyword> <Abstract>Privacy concerns with sensitive data are receiving increasing attention. In this thesis, we study local differential privacy (LDP) in interactive decentralized optimization. Comparing to central differential privacy (DP), where a centralized curator maintains the dataset, LDP is a stronger notion yet with industrial adoption, which allows data of an individual to be privatized before sharing. Consequently, more challenges are encountered to build efficient statistical analyzer in LDP setting. Towards practical decentralized optimization in LDP, we extend LDP into a more comprehensive notion which provides both worst and average case privacy guarantees. Accordingly, two approaches to sharpen utility-privacy tradeoff are proposed for the worst and the average, respectively: First, cryptographically incorporated with merely linear secret sharing, we show the privacy guarantee can be improved by a factor of [square root of] N&apos; where N&apos; amongst all N agents are semi-honest. Second, we take Alternating Direction Method of Multipliers (ADMM), and decentralized (stochastic) gradient descent(D(S)GD) as two concrete examples to propose a framework of first-order based optimization with random local aggregators. We prove such local randomization lead to the same utility guarantee but amplify average LDP by a constant, empirically around 30%. Thorough experiments support our theory.</Abstract> <Access xmlns="http://purl.org/coar/access_right" > </Access> </Publication>