Tuesday, May 6, 2025 | 2:00-3:30pm ET
2:00-3:00 Seminar
3:00-3:30 Questions and Collaboration

1430BD ISR-Thompson
426 Thompson St.

Register to attend

Abstract
This talk traces the concept of privacy from a late 19th-century legal right—spurred by tabloid harassment—to today's digital age challenges. It highlights Differential Privacy (DP),a cryptography-based method designed to balance data privacy and utility in a quantified way. However, despite DP's advancements and explicit warnings (e.g., Kifer and Machanavajjhala, 2011“No free lunch in data privacy”;Tschantz et al, 2022,“SoK:Differential privacy as a causal property”),misconceptions about its resistance to adversaries' prior knowledge persist. By revisiting Warner's (1965, JASA) randomized response mechanism, we argue that this misperception lies in treating data as static objects, rather than realizations of underlying, typically interdependent attributes or variables. We show how DP's effectiveness can falter when adversaries exploit interdependencies among individuals—similar to how quarantining only symptomatic individuals fails to stop an airborne disease. A holistic statistical perspective on joint modeling data is therefore as crucial for data privacy as for data analysis. (Joint work with James Bailie and Ruobin Gong.)

Bio:
Dr. Xiao-Li Meng is the Founding Editor-in-Chief of HDSR and the Whipple V. N. Jones Professor of Statistics, and the Founding Editor-in-Chief of Harvard Data Science Review, is well known for his depth and breadth in research, his innovation and passion in pedagogy, his vision and effectiveness in administration, as well as for his engaging and entertaining style as a speaker and writer. Meng was named the best statistician under the age of 40 by COPSS (Committee of Presidents of Statistical Societies) in 2001, and he is the recipient of numerous awards and honors for his more than 150 publications in at least a dozen theoretical and methodological areas, as well as in areas of pedagogy and professional development.