Privacy-preserving technologies 2022/23 spring

Lectures and practice sessions

Compared to the last year, the schedule of the course has been slightly reworked. The content has also been somewhat reworked.

The information in this page gives the materials for imminent and past lectures and practice sessions in this year, as well as references to similar lectures in past years. This information includes the lecture slides, as well as instructions to download and/or install something in your computers.

• Week 1 (08.02.2023, Dan Bogdanov):
  • Lecture: Course introduction. Introduction to privacy, data protection and Privacy Enhancing Technologies. Analysis of existing systems.
  • Seminar: Analysis of the intended and emergent effects of new services.
  • Homework: Analysis of the intended and emergent effects of new services.
• Week 2 (15.02.2023, Triin Siil): Legal status of privacy technologies.
• Week 3 (22.02.2023, Kristjan Krips): Attacks on privacy
  • Lecture notes: Attacks on privacy
  • Slides
  • Lab tasks
  • Recording of the lecture
  • Recording of the lab
  • OSINT homework
• Week 4 (01.03.2023, Raimundas Matulevičius): Designing privacy-preserving systems (part 1), focusing on GDPR.
  • Lecture slides
  • DPO tool (best with Google Chrome )
  • Used article
  • Homework: Designing privacy-preserving systems (DDL: 26.March)
• Week 5 (08.03.2023, Jan Willemson): introduction to cryptography, in non-technical way.
  • The slides are here and here.
• Week 6 (15.03.2023, Raimundas Matulevičius): Designing privacy-preserving systems (part 2), focusing on modelling of business processes using PETs.
  • Lecture slides
  • Pleak tool
  • Articles
    • Privacy-enhanced BPMN: enabling data privacy analysis in business processes models (link)
    • Business Process Privacy Analysis in PLEAK (link)
    • Multi-level privacy analysis of business processes: the PLEAK toolset (link)
  • Use case analysis slides
  • Use case model
  • Homework: Designing privacy-preserving systems (DDL: 26.March)
• Week 7 (22.03.2023, Liina Kamm): Introduction to data subject privacy enhancing techniques. Pseudonymisation. Anonymisation
  • Lecture slides
  • Recording of the 2nd half of the lecture
  • Recording of last year's lecture
• Week 8 (29.03.2023, Liina Kamm): Privacy-Preserving Federated Statistics and Machine Learning. Data synthesis.
  • Lecture slides
  • Recording of the lecture
• Week 9 (05.04.2023, Andre Ostrak, Liina Kamm): practice session on data anonymization (reshuffled due to illness).
  • There will be homework
  • Get the ARX tool here
  • Practice session example dataset
  • Recording of the lecture part 1
  • Recording of the lecture part 2
• Week 10 (12.04.2023, Alisa Pankova): Differential privacy.
  • See the recording from last year (slides), as well as a new recording (slides) clarifying certain aspects around rounding.
  • During the lecture, we will do some more practice-session-related tasks.
    • The tasks of the recorded, as well as the in-person session are here. There is the Python code implementing the queries, and the final version of the code that we ended up last year. The exercises are based on this data file.
  • There will be homework
  • Links to additional information:
    • Mathematical and technical background:
      • A short summary of SQL commands (SQL is used in the slides and in the exercises to describe database queries): https://www.cs.utexas.edu/~mitra/csFall2006/cs329/lectures/sql.html.
      • Probability density functions: https://www.khanacademy.org/math/statistics-probability/random-variables-stats-library/random-variables-continuous/v/probability-density-functions
    • Real-world applications of differential privacy
      • A brief overview of DP and some real-world use cases: https://research.aimultiple.com/differential-privacy/
      • More real-world use cases: https://desfontain.es/privacy/real-world-differential-privacy.html.
      • Impacts on the usefulness of the result (the example of US census data): https://eu.azcentral.com/story/news/local/arizona/2021/08/10/2020-census-data-differential-privacy/5541578001/
• Week 11 (19.04.2023, Pille Pullonen-Raudvere): privacy-preserving protocols for specific and general computational tasks. The slides are available here and the recording in BBB.
• Week 12 (26.04.2023, Riivo Talviste, Dan Bogdanov): Privacy-preserving computations. We will use Sharemind as technology for instruction.
  • Lecture slides, video
  • Lab instructions, video
  • Homework: Side-channel safe programming for MPC.
• Week 13 (03.05.2023, Nikita Snetkov and Peeter Laud): Cryptography for privacy preservation; privacy-preserving crypto primitives. Ring signatures, attribute-based encryption, zero-knowledge proofs, proxy re-encryption, etc. As non-technically as possible.
• Week 14 (10.05.2023, Nikita Snetkov, Peeter Laud): we will finish the last week's lecture on cryptography. In the remaining time, we will talk about Anonymous communication. This file is referenced from the slides of the latter.
• Week 15 (17.05.2023, Jan Willemson and Dan Bogdanov): example / exemplary use-cases for privacy-preserving cryptography.
  • Slides on internet voting (slides)
  • Slides on contact tracing (slides)
  • Slides on CSAM detection (slides)
• Week 16 (24.05.2023, Peeter Laud): Concluding seminars, and problem-solving. Group work that is similar to the individual work in the exam.
  • One of the groups worked on, an drew pictures about the population registry and power subsidies use-cases. They made some slides, too.
  • Another group worked on the building management use-case.
  • A couple of groups worked on the medical research use-case, resulting in this cooperatively-drawn picture.
  • A group based their discussion on the list of questions, indicating the following aspects to consider for the population registry and power subsidies use-cases.