Module Manual

Modul CS5076-KP12

Human-Centered Trustworthy AI (HumTrustAI)


1 Semester
Turnus of offer:

every summer semester
Credit points:

Course of studies, specific field and terms:
  • Master Artificial Intelligence 2023 (compulsory), Artificial Intelligence, 3rd and 4th semester
Classes and lectures:
  • TCHAI Lab (practical course, 2 SWS)
  • Cognition and Human-aware Interaction (lecture, 3 SWS)
  • Trustworthy AI (lecture, 3 SWS)
  • 90 Hours e-learning
  • 30 Hours work on project
  • 240 Hours private studies
Contents of teaching:
  • Human-Centered AI: Cognitive Modeling / Behavior Modeling / User and Group Modeling / Personalization / Cognitive Architectures / Human-aware Planning / Provably Beneficial AI / Ethics for AI Systems
  • Trustworthy AI: Guiding principles of Trustworthy AI: lawful, ethical and robust AI l Trustworthy Computing Basics: Security, Privacy, Dependability, Safety, Transparency, Explainability, Traceability, Accountability / De-anonymization methods using machine learning models / Mathematical notions for privacy-preserving machine learning methods / Privacy-preserving machine learning methods / Analysis of machine learned models (robustness check, explainability / Verification of machine learned models (statistical Testing, model checking) / Black-Box methods for extracting machine learning models (for economic reasons, for analysis, and for verification) / Attacks for manipulating machine learning models (adversarial examples, backdoors) Hardening of machine learning methods against manipulation methods / Robust machine learning methods against manipulation attacks / Secure and privacy-preserving distributed learning methods (privacy-preserving federated learning)
  • For all topics listed in the course content under the bullet points, students will be able to name the central ideas, define the relevant terms in each case, and explain how associated algorithms work using examples of applications.
Grading through:
  • portfolio exam
Responsible for this module:
  • N. Li, M. Lyu, D. Su, W. Yang: Differential Privacy: From Theory to Practice - Morgan Claypool, 2016
  • S. Farrel, S. Lewandowsky: Computational Modeling of Cognition and Behavior - Cambridge University Press, 2018
  • G. Marcus, E. Davis: Rebooting AI: Building Artificial Intelligence We Can Trust - Pantheon Books, 2019
  • S.J. Russell: Human Compatible: Artificial Intelligence and the Problem of Control - Penguin Books, 2020
  • M.H. ur Rehman, M.M. Gaber: Federated Learning Systems: Towards Next-Generation AI - Springer, 2021
  • C.S. Nam, J.-Y. Jung, S. Lee (Eds.): Human-Centered Artificial Intelligence: Research and Applications - Elsevier, 2022
  • B. Ammanath: Trustworthy AI: A Business Guide for Navigating Trust and Ethics in AI - Wiley, 2022
  • offered only in English

Prerequisites for attending the module:
- None (The competencies of the modules listed under 'Requires' are needed for this module, but are not a formal prerequisite)

Prerequisites for the exam:
- 50% of online quiz points

Module exam(s):
CS5076-L1: Human-Centered Trustworthy AI portfolio exam for a total of 100 points, divided as follows:
- 50 points for an e-test (oral or written).
- 50 points for a project presentation

Letzte Änderung: