Completed
Optimized Private Gradient Descent
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Cynthia Dwork: The Mathematics of Privacy
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Fundamental Law of Info Reconstruction • Overly accurate" estimates of too many" statistics is
- 3 Statistics 'Feel Private
- 4 Privacy Preserving Data Analysis
- 5 Differential Privacy M gives e-differential privacy if for all pairs of adjacent data
- 6 Some Properties of Differential Privacy
- 7 The Laplace Mechanism
- 8 The Privacy Loss Random Variable
- 9 Advanced Composition Theorem • Recall privacy loss is sometimes negative -- there is cancellation
- 10 Gaussian Mechanism
- 11 Concentrated Differential Privacy
- 12 Privacy Amplification via Subsampling
- 13 (6,8)-DP Projected Gradient Descent
- 14 Optimized Private Gradient Descent
- 15 Creative Privacy Accounting Thought Experiment: Consider two steps of Noisy-SGD with fixed sample order
- 16 Amplification by Secrecy of the Journey
- 17 Challenge
- 18 Crucial Definition
- 19 "Shift" Calculus
