Mattia Setzu

Mattia Setzu is a Research Fellow at the Department of Computer Science, University of Pisa, where he also pursues his PhD. His work focuses on Explainable Artificial Intelligence (XAI), particularly on developing methods to interpret and explain the decisions made by black box machine learning models.

His educational background includes:

• Bachelor's degree in Computer Science from the University of Cagliari (2016) with 103/110
• Master's degree in Computer Science from the University of Pisa (2018) with 110/110 cum laude

Setzu's research centers on the explainability problem in artificial intelligence, addressing what he refers to as the "black box syndrome" that machine learning models suffer from. His quote "To be human is to wonder. To go beyond the empirical data and grasp the essence of things is the cornerstone of human understanding and allows us to be active agents in the world" reflects his philosophical approach to explainable AI. His methodological evolution spans from semantic web applications for source code analysis to sophisticated frameworks for generating global explanations from local ones using clustering approaches, topology-aware algorithms, and Bayesian techniques.

His publication record demonstrates a clear progression in XAI research, with recent work connecting explainability methods to critical applications in pandemic response. This interdisciplinary approach shows how his core research on making AI systems transparent has practical implications in high-stakes domains like public health and social policy.

Setzu contributes to major research initiatives:

• XAI - Science and technology for the eXplanation of AI decision making
• TAILOR - Foundations of Trustworthy AI - Integrating Reasoning, Learning and Optimization

His collaborative work extends across multiple domains, frequently appearing as co-author with researchers from diverse backgrounds including public health, social science, and traditional computer science disciplines. This network reflects the interdisciplinary nature of modern AI research and the growing importance of explainability across application domains.