Methods and Considerations for Using Differential Privacy in Government Statistics

Mon, Aug 4, 2:00 PM - 3:50 PM Room CC-101D Music City Center

Mikaela MeyerOrganizerZachary TernerOrganizerZachary TernerChair

Government Statistics Section co: Survey Research Methods Section Applied

5 Presentations

Slowly Scaling Per-Record Differential Privacy

2:05 PM - 2:25 PM

Mikaela Meyer

Co-authors: Brian Finley (US Census Bureau), Anthony Caruso (U.S. Census Bureau), Justin Doty (U.S. Census Bureau), Ashwin Machanavajjhala, David Pujol (Tumult Labs), William Sexton (Tumult Labs), Zachary Terner (The MITRE Corporation)

Best Linear Unbiased Estimate from Privatized Histograms

2:25 PM - 2:45 PM

Jordan Awan (Purdue University)

Co-authors: Adam Edwards (The Mitre Corporation), Paul Bartholomew (The Mitre Corporation), Andrew Sillers (The MITRE Corporation)

WITHDRAWN Exploring Statistical Disclosure Limitation Techniques for Agricultural Data Using Robust Metrics for Risk and Utility

2:45 PM - 3:05 PM

Co-authors: Luca Sartore (National Institute of Statistical Sciences), Valbona Bejleri (United States Department of Agriculture – National Agricultural Statistics Service)

Differential Privacy and the Survey Data Pipeline

3:05 PM - 3:25 PM

Joerg Drechsler (Institute for Employment Research)

Co-authors: James Bailie (Harvard University)

Implementing PETs for Government Data: Context and Considerations

3:25 PM - 3:45 PM

Meghan Stuessy (US Congressional Research Service)

Co-authors: Meghan Stuessy (US Congressional Research Service), Jake Pasner (Georgetown University), Christine Task (Knexus Research Corporation)