We developed hands-on active-learning sessions to accompany a theoretical introductory course on differential privacy. The course is designed for undergraduate and graduate students in computer science.

As a teaching assistant and course co-developer, I wrote exercises, developed course materials, taught tutorials, and graded coursework.

Course outline:

The course introduced the foundations of data privacy and differential privacy through theoretical concepts, privacy attacks, and core algorithmic tools. The tutorials and exercises were designed to help students actively work with definitions, mechanisms, and proof techniques.

Course Topics:

• Privacy attacks and privacy notions

• Basic properties of differential privacy

• Core differential privacy mechanisms:
  • Randomized Response
  • Laplace mechanism
  • ReportNoisyMax
  • AboveThreshold
  • Sparse Vector
  • Exponential Mechanism
• Synthetic databases:
  • SmallDB
  • MWEM and online MWEM
• Advanced topics:
  • Propose-test-release and smooth sensitivity
  • Local DP
  • Amplification by shuffling
  • Differentially private machine learning