The EDA group at LIACS develops algorithms and theory that help people discover what matters in complex data, understand cause and effect, and make trustworthy predictions and decisions.

Our research has two main, complementary themes: discovering what matters and from causes to decisions. Across both themes, our work builds on machine learning and statistics. Information theory and the minimum description length (MDL) principle, probabilistic and causal graphical models, and stochastic optimisation give us principled ways to discover structure, reason about interventions, balance goodness of fit against complexity, and make accurate predictions and effective decisions.

Many of our fundamental research questions arise from collaborations in health and life sciences, people and society, and industry and engineering. Practical challenges expose limitations of existing methods and inspire new algorithms and theory. In turn, these methods can help our partners obtain clearer explanations, more reliable predictions, and new domain knowledge.