A session at the 2019 Joint Statistical Meetings, sponsored by the ASA Committee on Privacy and Confidentiality.

Lars Vilhuber

Aleksandra Slavković

With the ever-increasing amount of data collected everyday, data confidentiality is more and more at risk. Many of the traditional approaches to statistical disclosure control are no longer sufficient to protect the confidentiality of the data. Formal privacy guarantees - provable privacy guarantees that hold regardless of assumed knowledge and attack strategy of a malicious user - are becoming increasingly important for large producers of statistics, such as national statistical agencies or large private companies. These organizations need to design and engineer systems with effective formally private disclosure limitation systems. This session brings together experts who have developed, proposed, and implemented formal privacy models such as variants of differential privacy in various large organizations. The presentations will inform attendees of challenges that were identified, how they were met, and the outlook for future implementations.

• Simson L. Garfinkel, U.S. Census Bureau;
• Juan Lavista Ferres, Microsoft;
• Ilya Mironov, Google;
• Shiva Kasiviswanathan, Amazon

Simson Garfinkel is the Senior Computer Scientist for Confidentiality and Data Access at the US Census Bureau. He holds seven US patents and has published more than 50 research articles in computer security and digital forensics. He is a fellow of the Association for Computing Machinery (ACM) and the Institute of Electrical and Electronics Engineers (IEEE), and a member of the National Association of Science Writers. His most recent book is The Computer Book, which features 250 chronologically arranged milestones in the history of computing. As a journalist, he has written about science, technology, and technology policy in the popular press since 1983, and has won several national journalism awards.

Garfinkel received three Bachelor of Science degrees from MIT in 1987, a Master's of Science in Journalism from Columbia University in 1988, and a Ph.D. in Computer Science from MIT in 2005.

(source)

Presentation

Juan Miguel Lavista is currently the Senior Director of Data Science for Microsoft AI For Good where he works with a team of data scientists in AI, Machine Learning and statistical modeling, working across Microsoft AI for Good initiatives. Juan joined Microsoft in 2009 to work for the Microsoft Experimentation Platform (EXP) where he designed and ran randomized control experiments across different Microsoft groups. Juan also worked as part of the Bing Data Mining team, where he led a group applying data mining, machine learning, statistical modeling and online experimentation at a large scale as well as providing data services for Bing.

Juan is involved in working to define the data science discipline within Microsoft, and is currently the editor of the Microsoft Journal of Applied Research (MSJAR).

Before joining Microsoft, Juan was the CTO and co-founder of alerts.com. Previously, he spent 6 years in Washington working at the InterAmerican Development Bank applying data science to understand the impact of programs for reducing poverty and inequality in Latin-America and the Caribbean. Juan has two computer science degrees from the Catholic University in Uruguay, and a graduate degree in Data Mining from Johns Hopkins University.

Presentation

Ilya Mironov is a computer scientist working in privacy, cryptography, and machine learning. While in Microsoft Research - Silicon Valley (2003-2014), he became active in the community centered around the definition of differential privacy, studying its applications, limitations, and relaxations. In Google, he worked on large-scale deployment of differentially private mechanisms across several product areas.

Presentation

Shiva Kasiviswanathan is a Senior Applied Scientist at Amazon. He graduated with a PhD in Computer Science from Pennsylvania State University in 2008, and has since worked as a researcher at various research labs. His research interests span differential privacy and machine learning. His research contributions have been successfully implemented into various commercial products at IBM, GE, Samsung, and Amazon. He has published more than 50 peer-reviewed publications, including top conference venues such as ICML, NIPS, KDD, STOC, FOCS, and VLDB.