Differential privacy with bounded priors: Reconciling utility and privacy in genome-wide association studies

dc.citation.epage1297en_US
dc.citation.spage1286en_US
dc.contributor.authorTramèr, F.en_US
dc.contributor.authorHuang, Z.en_US
dc.contributor.authorHubaux J.-P.en_US
dc.contributor.authorAyday, Ermanen_US
dc.coverage.spatialDenver, Colorado, USA
dc.date.accessioned2016-02-08T12:16:42Z
dc.date.available2016-02-08T12:16:42Z
dc.date.issued2015-10en_US
dc.departmentDepartment of Computer Engineeringen_US
dc.descriptionConference name: CCS '15 Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security
dc.descriptionDate of Conference: 12 - 16 October, 2015
dc.description.abstractDifferential privacy (DP) has become widely accepted as a rigorous definition of data privacy, with stronger privacy guarantees than traditional statistical methods. However, recent studies have shown that for reasonable privacy budgets, differential privacy significantly affects the expected utility. Many alternative privacy notions which aim at relaxing DP have since been proposed, with the hope of providing a better tradeoff between privacy and utility. At CCS'13, Li et al. introduced the membership privacy framework, wherein they aim at protecting against set membership disclosure by adversaries whose prior knowledge is captured by a family of probability distributions. In the context of this framework, we investigate a relaxation of DP, by considering prior distributions that capture more reasonable amounts of background knowledge. We show that for different privacy budgets, DP can be used to achieve membership privacy for various adversarial settings, thus leading to an interesting tradeoff between privacy guarantees and utility. We re-evaluate methods for releasing differentially private χ2-statistics in genome-wide association studies and show that we can achieve a higher utility than in previous works, while still guaranteeing membership privacy in a relevant adversarial setting. © 2015 ACM.en_US
dc.description.provenanceMade available in DSpace on 2016-02-08T12:16:42Z (GMT). No. of bitstreams: 1 bilkent-research-paper.pdf: 70227 bytes, checksum: 26e812c6f5156f83f0e77b261a471b5a (MD5) Previous issue date: 2015en
dc.identifier.doi10.1145/2810103.2813610en_US
dc.identifier.urihttp://hdl.handle.net/11693/28296
dc.language.isoEnglishen_US
dc.publisherACM
dc.relation.isversionofhttp://dx.doi.org/10.1145/2810103.2813610en_US
dc.source.titleProceedings of the ACM Conference on Computer and Communications Securityen_US
dc.subjectData-driven medicineen_US
dc.subjectDifferential privacyen_US
dc.subjectGenomic privacyen_US
dc.subjectGWASen_US
dc.subjectMembership privacyen_US
dc.subjectBudget controlen_US
dc.subjectGenesen_US
dc.subjectKnowledge managementen_US
dc.subjectProbability distributionsen_US
dc.subjectBack-ground knowledgeen_US
dc.subjectData drivenen_US
dc.subjectDifferential privaciesen_US
dc.subjectExpected utilityen_US
dc.subjectGenome-wide association studiesen_US
dc.subjectPrior distributionen_US
dc.subjectPrivacy frameworksen_US
dc.subjectData privacyen_US
dc.titleDifferential privacy with bounded priors: Reconciling utility and privacy in genome-wide association studiesen_US
dc.typeConference Paperen_US

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Differential privacy with bounded priors Reconciling utility and privacy in genome-wide association studies.pdf
Size:
793.41 KB
Format:
Adobe Portable Document Format
Description:
Full printable version