Senior Research Scientist
Lakera AI
Previously: Associate Professor of Statistics
Copenhagen Causality Lab
Department of Mathematical Sciences, University of Copenhagen
My research focuses on statistical methodology for modern data applications, that either involve large models, complex data structures or both. Developing statistical tools that help us understand the underlying mechanisms in these applications is becoming more important as their complexity and sophistication increases.
Much of my previous work uses some notion of causality and stability to model and infer parts of these mechanisms. At Lakera I work on developing methods to keep generative AI applications safe and secure. This involves a lot of exciting research, including on large pre-trained models. Understanding the inner workings of these models is a crucial step in developing effective defenses.
Keywords: applied biology, causality, dynamical systems, independence testing, kernel methods, large pre-trained models, time series