Trustworthy and Resposible Artificial Intelligence

Recommended Prerequisites

Programming, mathematical, artificial intelligence and machine learning competencies.

Teaching Methods

The theoretical classes consist of a detailed presentation of concepts, principles, and fundamental theories of trustworthy and responsible Artificial Intelligence.

In the laboratory classes, a Project-Based Learning approach is adopted, focused on acquiring skills through practical work, involving three components: (i) analysis of works on trustworthy and responsible Artificial Intelligence described in the literature, (ii) implementation, and (iii) writing a scientific paper that describes the implementation work and can be presented and defended.

Learning Outcomes

The course aims to provide students with a comprehensive understanding of trustworthy and responsible AI, equip them with critical thinking skills, and enable them to design and deploy AI systems that align with ethical principles, fairness, transparency, accountability, privacy, security, and sustainability.

Work Placement(s)

No

Syllabus

 

1. Introduction to Responsible and Trustworthy AI

 

2. Human Agency and Oversight

2.1 Collaborative AI

2.2 Human-in-the-Loop vs Human-in-Command vs. Human-out-of-the Loop

2.3 Apprenticeship Learning

2.4 User modelling and Personalization

 

3. Transparency and Interpretability/Explainability in AI systems

3.1 Motivation

3.2 Challenges

3.3 Approaches

 

4. Diversity, Non-discrimination and Fairness in AI systems

4.1 Motivation

4.2 Challenges

4.3 Approaches

 

5. Accountability in AI Systems

5.1 Motivation

5.2 Challenges

5.3 Approaches

 

6. Robustness and Safety in AI Systems

6.1 Motivation

6.2 Challenges

6.3 Approaches

 

7. Privacy and data governance in AI Systems

7.1 Motivation

7.2 Challenges

7.3 Approaches

 

8. Societal and Environmental Well-being in AI Systems

8.1 Motivation

8.2 Challenges

8.3 Approaches

 

9. Ethics and Morality in AI Systems

9.1 Motivation

9.2 Challenges

9.3 Approaches

 

10. Frameworks and Guidelines

10.1 Establishement in organizations

10.2 Compliance

 

11. Applications

Assessment Methods

Assessment
Exam: 40.0%
Project: 60.0%

Bibliography

European Commission High-Level Expert Group on AI, Ethics guidelines for trustworthy AI (2019).

 

European Union, Proposal for a Regulation of the European Parliament and of the Council Laying down harmonised rules on Artificial Intelligence (Artificial Intelligence Act) and amending certain Union Legislative Acts. COM/2021/206 final (2021).

 

UNESCO, Recommendation on the ethics of artificial intelligence, Digital Library UNESDOC (2020). URL en.unesco.org

 

Molnar, C. Interpretable machine learning. A Guide for Making Black Box Models Explainable, 2019. https://christophm.github.io/interpretable-ml-book/.

 

Russell, S. (2019). Human compatible: Artificial intelligence and the problem of control. Viking.

 

Dignum, V. (2021). Responsible artificial intelligence: How to develop and use AI in a responsible way. Springer.

Dubber, M.D., Pasquale, F. and Das, S. (Eds.) (2020). The Oxford handbook of ethics of AI. Oxford University Press.