In her inaugural lecture, Katrien Antonio will start by explaining some typical actuarial predictive problems, followed by an exploration of the landscape of learning methods for actuarial outcome variables. These variables include the frequency and severity of an event and the time-to-an-event. She will then subsequently explain how actuaries harvest insights from fine-grained input data of different types, collected from different sources, and how these insights are used by insurers. Operating in a highly regulated industry with high stakes decisions, actuaries aim to balance predictive accuracy on the one hand, and transparency and fairness of their learning models on the other hand. This calls for reflections on and a roadmap for responsible actuarial learning. And this is the final topic Antonio will address in her lecture.

Prof. K. Antonio, professor of Actuarial Data Science: Responsible actuarial learning.