Completed
The basic idea of differential privacy: Uncertainty (noise) protects privacy
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Using Apache Spark and Differential Privacy for 2020 Census Data Protection
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Abstract
- 3 Outline
- 4 Privacy and the Decennial Census
- 5 2010 Census: Summary of Publications (approximate counts)
- 6 We performed a database reconstruct and re-identification attack for all 308.745538 people in the 2010 Census
- 7 The basic idea of differential privacy: Uncertainty (noise) protects privacy
- 8 The Census Bureau is using differential privacy for the 2020 Census.
- 9 How much noise do we add? That's a policy decision.
- 10 We planned to create a Disclosure Avoidance System that dropped into the Census production system.
- 11 The Disclosure Avoidance System allows the Census Bureau to enforce global confidentiality protections
- 12 Our DP mechanism protects histograms of person types. Census "block"
- 13 Running the block-by-block algorithm with spark
- 14 In 2018 we invented the TopDown Algorithm (TDA)
- 15 Key challenges in monitoring spark
- 16 We created our own monitoring framework
- 17 Cluster List
- 18 Each DAS run is a "mission"
- 19 Mission Report
- 20 System Load
- 21 Free Memory
- 22 In Summary
