Rianne de Heide

Rianne de Heide is a Researcher in the Machine Learning department at Centrum Wiskunde & Informatica (CWI) in Amsterdam, The Netherlands. Her work focuses on theoretical aspects of statistical learning, particularly in the areas of Bayesian inference and hypothesis testing.

Dr. de Heide's research interests span several interconnected domains in statistical theory and machine learning:

• Bayesian statistics and inference
• Hypothesis testing and statistical validity
• E-values and anytime-valid testing
• Group invariance in statistical models
• Safe Bayesian learning under misspecification
• Multi-armed bandit problems

Her recent publications demonstrate a strong focus on developing theoretically sound statistical methods that maintain validity under flexible conditions. A significant portion of her work explores 'safe testing' frameworks that remain valid regardless of when testing is stopped, addressing a long-standing challenge in statistical practice. Her research bridges theoretical statistics with practical applications in machine learning, with publications in top journals including the Annals of Statistics and the Journal of the Royal Statistical Society Series B.

Dr. de Heide has received recognition for her work, including:

• Willem R. van Zwet Award (2022)

Her research is supported by the project 'Safe Bayesian Inference: A Theory of Misspecification based on Statistical Learning,' funded by The Netherlands Organisation for Scientific Research (NWO). She collaborates extensively with Peter Grünwald, Wouter Koolen-Wijkstra, and other researchers in the Machine Learning group at CWI.