The Geometry-Topology Seminar continues with Shiva Kasiviswanathan of IBM. As usual, the talk is at 3:30 in Wachman 617.
Title: The Price of Privately Releasing Contingency Tables and the Spectra of Random Matrices with Correlated Rows
Abstract: Contingency (marginal) tables are the method of choice of government agencies for releasing statistical summaries of categorical data. However, if the contingency tables are released exactly, one can reconstruct the individual entries of the data by solving a system of equations. In this talk, we give tight bounds on how much distortion (noise) is necessary in these tables to provide privacy guarantees when the data being summarized is sensitive. Our investigation also leads to new results on the spectra of random matrices with correlated rows.
Based on joint work with Mark Rudelson, Jonathan Ullman, and Adam Smith.