Oberseminar über Differential Privacy
(MA XX: Modul XX; MSc Mod XX: Modul X)
SoSe 2019
| Groups | Time | Room | Start |
|---|---|---|---|
| H. Dette, J. Lederer, A. May | Tue 11:00-12:00 | Raum IA 1/53 | April 23 |
Literatur
We use the excellent lecture notes of
Aaron Roth, Adam Smith, The Algorithmic Foundations of Adaptive Data Analysis, 2017.
| Date | Speaker | Topic | Lecture |
|---|---|---|---|
| 23.04. | Önder Askin | Introduction, tail bounds for one statistical query | Lecture 1 |
| 30.04. | Tim Kutta | Many nonadaptive queries, uniform convergence, and stochastic optimization. | Lecture 2 Folien |
| 07.05. | Shih-Ting Huang | Interactive data analysis, and the dangers of adaptivity. | Lecture 3 Folien |
| 14.05. | Mahsa Taheri | Description length bounds I: Compressibility and generalization | Lecture 4 |
| 21.05. | Alexander Helm | Description length bounds II: Rounding, AboveThreshold and Ladder algorithm | Lecture 5Folien |
| 28.05. | Vincent Göbels | The “median mechanism”,“reusable holdout” with description length bounds | Lecture 6 Folien |
| 04.06. | Lars Schlieper | Algorithmic and distributional stability | Lecture 7, Chapter 1-2.1 |
| 18.06. | Eftychia Solea | The monitor argument and adaptive statistical queries | Lecture 8, Chapter 2.2-3 |
| 02.07. | Claire Delaplace | Expected error bounds for the Gaussian mechanism | Lecture 9, Chapter 4-6 |