Date(s) - 03/06/2017
10:30 am - 11:30 am
Privacy is an important constraint that algorithms must satisfy when analyzing sensitive data from individuals. Differential privacy, a provable property of certain algorithms, has arisen as a gold standard for exploring the tradeoff between the privacy ensured to individuals and the utility of the statistical insights mined from the data. Differential privacy is starting to see adoption in many commercial (e.g., Google and Apple) and government entities (e.g., US Census) for collecting and sharing sensitive user data.
In today’s talk I will highlight some of the open challenges in designing differentially private algorithms for emerging applications, and highlight research form our group that try to address these challenges. In particular I will describe our recent work on modernizing the data publication process for a US Census Bureau data product, called LODES/OnTheMap. In this work, we identified legal statutes and their current interpretations that regulate the publication of these data, formulated these requirements mathematically, and designed algorithms for releasing tabular summaries that provably ensured these privacy requirements. Our solutions are able to release summaries of the data with error comparable or even better than current releases (which are not provably private), for reasonable settings of privacy parameters. Joint work with Sam Haney (Duke), John Abowd, Matthew Graham, Mark Kutzbach (US Census Bureau) and Lars Vilhuber (Cornell).