I work on Bayesian inference, mostly for robots that need to know where they are. Right now that means real-time filtering, underwater tracking and SLAM, and checking whether the uncertainty a filter reports can be trusted. Often it can't (see our odometry paper).

I'm a postdoc in the SQUARE group at the IT University of Copenhagen, on a DDSA fellowship. I did my PhD with Thomas Hamelryck at the University of Copenhagen on Stein mixture inference.

News

Selected Publications

Ola Rønning, Usama Saqib, Andrzej Wąsowski (2026). You Should Be Properly Scoring Your Odometry. arXiv:2609.25900. Preprint
Ola Rønning, Eric Nalisnick, Christophe Ley, Thomas Hamelryck (2025). ELBOing Stein: Variational Bayes with Stein Mixture Inference. International Conference on Learning Representation.
Ola Rønning, Christophe Ley, Kanti V. Mardia, Thomas Hamelryck (2021). Time-efficient Bayesian Inference for a (Skewed) Von Mises Distribution on the Torus in a Deep Probabilistic Programming Language. IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems.