BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Talks.cam//talks.cam.ac.uk// X-WR-CALNAME:Talks.cam BEGIN:VEVENT SUMMARY:PSI : a Private data Sharing Interface - Marco Gaboardi (Universit y of Dundee) DTSTART:20161208T113000Z DTEND:20161208T120000Z UID:TALK69391@talks.cam.ac.uk CONTACT:INI IT DESCRIPTION:Co-authors: \;James Honaker \;(Harvard University)&nbs p\;\, Gary King \;(Harvard University) \;\, Jack Murtagh \;(Ha rvard University) \;\, Kobbi Nissim \;(Ben-Gurion \;University and CRCS Harvard University) \;\, Jonathan Ullman \;(Northeastern University) \;\, Salil Vadhan \;(Harvard University)

We provide an overview of the design of PSI (&ldquo\;a Private d ata Sharing Interface&rdquo\;)\, a system we are developing to enable rese archers in the social sciences and other fields to share and explore priva cy-sensitive datasets with the strong privacy protections of differential privacy.
PSI is designed so that none of its users need expertise in pr ivacy\, computer science\, or statistics. PSI enables them to make informe d decisions about the appropriate use of differential privacy\, the settin g of privacy parameters\, the partitioning of a privacy budget across diff erent statistics\, and the interpretation of errors introduced for privacy .
Additionally\, PSI is designed to be integrated with existing and wid ely used data repository infrastructures as part of a broader collection o f mechanisms for the handling of privacy-sensitive data\, including an app roval process for accessing raw data (e.g. through IRB review)\, access co ntrol\, and secure storage.
Its initial set of differentially pri vate algorithms were chosen to include statistics that have wide use in th e social sciences\, and are integrated with existing statistical software designed for modeling\, interpreting\, and exploring social science data.< br>
Related Links LOCATION:Seminar Room 1\, Newton Institute END:VEVENT END:VCALENDAR