Dagstuhl-Seminar 27032
Data Sharing and Differential Privacy
( 17. Jan – 21. Jan, 2027 )
Permalink
Bitte benutzen Sie folgende Kurz-Url zum Verlinken dieser Seite:
https://www.dagstuhl.de/27032
Organisatoren
- Monika Henzinger (IST Austria - Klosterneuburg, AT)
- Tamalika Mukherjee (MPI-SP - Bochum, DE)
- Rasmus Pagh (University of Copenhagen, DK)
- Sergei Vassilvitskii (Google - New York, US)
Kontakt
- Michael Gerke (für wissenschaftliche Fragen)
- Simone Schilke (für administrative Fragen)
Differential privacy (DP) is the leading framework for publishing useful statistics with rigorous privacy guarantees. This Dagstuhl Seminar concentrates on DP for data sharing and statistical data release, where the goal is to approximate well-defined functions of data (e.g., counts, histograms, marginals, and time series), for example via synthetic data (artificial datasets that preserve statistical properties). This focus is of interest for public agencies and regulators, as well as other organizations that need to share information derived from datasets while respecting privacy requirements.
Major deployments of DP, such as the U.S. Census Bureau's 2020 TopDown system, and EU regulatory drivers (notably the 2022 Digital Markets Act on sharing search-related data with anonymisation) raise questions that are both theoretical and urgent in areas such as: tight privacy accounting under composition; adaptive (workload-aware) mechanisms for structured releases; reconciliation under policy invariants; principled utility metrics; and streaming DP for continuous publication. This seminar will be research-oriented, deepening theory while learning from and being motivated by practice. It aims to bring together researchers in design and analysis of differentially private algorithms with practitioners from statistical agencies and industry.
Creative Commons BY 4.0
Monika Henzinger, Tamalika Mukherjee, Rasmus Pagh, and Sergei Vassilvitskii
This seminar qualifies for Dagstuhl's LZI Junior Researchers program. Schloss Dagstuhl wishes to enable the participation of junior scientists with a specialisation fitting for this Dagstuhl Seminar, even if they are not on the radar of the organizers. Applications by outstanding junior scientists are possible until Friday, September 11, 2026.
Klassifikation
- Cryptography and Security
- Data Structures and Algorithms
- Databases
Schlagworte
- differential privacy
- statistical data release
- synthetic data
- streaming DP
- data sharing
- query workloads
- privacy accounting
- EU DMA 6(11)
