Machine Learning: Privacy, Regulations and Ethical Issues (SS 2023)

Module Description

Machine Learning is becoming increasingly important in daily life applications such as self-driving cars or communication assistants. These applications require large amounts of data, which is why many countries have introduced regulations like the EU GDPR to govern data usage. Alongside privacy concerns, ethical questions regarding fair AI also arise.

The challenge is to open new markets while meeting ethical, privacy, and regulatory requirements. Various privacy-preserving machine learning technologies have emerged to protect models from attacks attempting to reveal data, training features, or the algorithm itself. Additionally, fairness approaches help define rules for ethical AI applications.

Objectives

• Understand privacy issues and possible solutions for Machine Learning applications
• Understand ethical issues and possible solutions for Machine Learning applications
• Understand the aim and need for Machine Learning regulations

Topics

• Analysis of different Machine Learning applications
• Ethical issues in Machine Learning
• Fairness and data bias
• Privacy Preserving Machine Learning Techniques
• Machine Learning in 6G

Literature

• Act, A. I. (2021). Proposal for a regulation of the European Parliament and the Council laying down harmonised rules on Artificial Intelligence (Artificial Intelligence Act). EUR-Lex-52021PC0206.
• Su, J., Vargas, D. V., & Sakurai, K. (2019). One pixel attack for fooling deep neural networks. IEEE Transactions on Evolutionary Computation, 23(5), 828-841.
• Smuha, N. A. (2019). The EU approach to ethics guidelines for trustworthy artificial intelligence. Computer Law Review International, 20(4), 97-106.
• Tronnier, F., Pape, S., Löbner, S., & Rannenberg, K. (2022). A Discussion on Ethical Cybersecurity Issues in Digital Service Chains. In Cybersecurity of Digital Service Chains (pp. 222-256). Springer, Cham.

Learning Goals and Competencies

• Ability to understand and perform a systematic literature review (LGB-1)
• Basic understanding of different Machine Learning approaches (LGB-5)
• Basic understanding of Regulation and Privacy in Machine Learning (LGB-3)
• Basic understanding of Ethical Issues in Machine Learning (LGB-4)
• Demonstrate good writing and presentation skills (LGB-7)
• Demonstrate good organisational skills and collaboration in working in groups (LGB-6)

Stage 1: Application

Deadline: 20 March 2023 – 03 April 2023

Module Application via QIS: My Functions > Lectures occupy/sign off

Stage 2: Exam Registration and Withdrawal

Deadline: 06 April 2023 – 19 April 2023

Exam Registration and Withdrawal via QIS: My Functions > Administration of exams

An acceptance in the application procedure entitles students to register for the seminar but does not replace an exam registration. Without an exam registration in stage 2, the seminar claim from stage 1 expires.

Note: There is no assignment of available capacities for this seminar. Exam registration is only possible within the deadline and with prior acceptance.

For Exchange Students

Module application and exam registration are not possible via QIS. Exchange students must register or withdraw using a form within the exam registration and withdrawal deadline (not the application deadline). Forms are available on the Faculty’s International Office website.

Assessment

Successful completion requires a term paper with presentation and regular attendance.

Specific topics will be introduced during the kick-off, and methodologies will be discussed before topic allocation. Students must carefully work through the methodology of their topic.

Exam dates and retake exams will be published on this website at the beginning of the semester.

Please check your student emails (@stud.uni-frankfurt.de) regularly and stay updated with