Would you trust a computer that says “no” without explaining why? In the minor Explainable AI (XAI), you will learn how to make complex algorithms transparent and fair. Combat bias in data and learn how to translate complex decision-making into understandable visualizations. Become the expert who bridges the gap between advanced technology and human trust. Help build responsible AI that is not only smart, but also fair!

The Explainable AI (XAI) minor focuses on one of the greatest challenges in modern technology: making artificial intelligence understandable and responsible. As AI systems become more powerful, they often become less transparent. In this minor, you will learn how we can restore trust in technology by opening up 'black-box' systems and making them understandable to humans.

 The program is structured around the following core themes:

 1. Foundations of XAI. We start with the meaning and necessity of Explainable AI. Why isn't it enough for a model to be accurate? You will learn about the trade-off between a model's complexity and its interpretability, and why sectors such as healthcare, banking, and government are crying out for transparent solutions.

 2. Inherent Interpretability. Not all AI needs to be a mystery. We investigate models that are inherently understandable. You will delve into linear models, decision trees, and rule-based models. You will learn how to design these models so that humans can immediately see which factors led to a particular outcome.

 3. Opening the Black Box (Model-Agnostic Methods). Sometimes a complex model (such as a deep neural network) is unavoidable. In this module, you'll learn techniques that can explain any "black box." You'll work hands-on with methods like LIME (Local Interpretable Model-Agnostic Explanations) and SHAP (Shapley Additive Explanations). These allow you to pinpoint exactly which variables were decisive in a specific prediction.

 4. Visualization & Human Perception. An explanation is only useful if a human understands it. You'll learn how to translate complex data and model decisions into effective visualizations. We'll look at the psychology of trust and how to present the right information to different audiences, from data scientists to policymakers.

 5. Ethics, Law, and Justice. XAI doesn't stand alone; it's closely linked to responsibility. We'll cover crucial themes such as bias (prejudices in data) and fairness. You will learn how transparency helps detect discrimination by algorithms and how to comply with legal requirements such as the 'right to an explanation' under the GDPR and the EU AI Act

 6. Human-AI Collaboration. AI is at its best when it enhances humans. You will investigate how transparent systems improve collaboration between humans and machines. How do you ensure that a doctor trusts AI advice? How do you prevent overreliance (blind trust)? You will develop strategies for a healthy interaction between human expertise and artificial intelligence.

 7. Practice & Case Studies. During the minor, you will directly apply your knowledge to real-world challenges.

After completing the Explainable AI minor, students will be able to:

• Argue the need for and principles of transparency and explainability within AI and machine learning.
• Evaluate and implement inherently interpretable models (such as linear models and decision trees).
• Apply model-agnostic methods to analyze the decision-making process of complex black-box models.
• Use advanced visualization techniques to make model predictions understandable for both experts and end users.
• Test AI systems against ethical and legal frameworks regarding fairness, bias, and accountability.
• Design strategies for Human-AI collaboration that build trust through transparency.
• Translate theory into practice by applying XAI methodologies to real-world business cases.

Target groups: all study programmes

No specific prior education is required for the minor. However, basic knowledge of mathematics (level havo A or B) and programming is assumed.

Description of meetings: Lectures, discussions, workshops and an ongoing project. 
Attendance required during assessment moments and collaboration in project groups.

Contact hours spread throughout the week during the day; 10 to 15 hours per week

Performance profile, 15 EC, minimum 5,5: The student delivers performance based on established performance indicators. Work attitude (behaviour), professional products (skills) and written test (knowledge) will be assessed.

Locations: Maastricht Paul-Henri Spaaklaan and Heerlen Brightlands Campus