Posters at the EWRL 2025 — All Good Things Come in Threes
This September, two members of our lab participated in the European Workshop on Reinforcement Learning (EWRL), one of the leading international gatherings for the RL community. The 2025 edition, hosted at the University of Tübingen in Germany, brought together researchers from around the world to share cutting-edge work and discuss open challenges in reinforcement learning. Tübingen, with its rich history and charming streets, provided a perfect backdrop for both scientific exchange and social activities, including a memorable dinner in the historic Hohentübingen Castle’s Museum of Ancient Cultures.
The workshop was attended by two ELLIS PhD candidates from our labs: Maris Galesloot, a doctoral researcher at Radboud University (and member of our partner lab in Nijmegen, also led by Prof. Nils Jansen), and Joshua Wendland, a doctoral researcher at Ruhr University Bochum.
Our group presented three posters at the workshop:
On Evaluating Policies for Robust POMDPs (Merlijn Krale, Eline M. Bovy, Maris F. L. Galesloot, Thiago D. Simão, Nils Jansen)
Missingness-MDPs: Bridging the Theory of Missing Data and POMDPs (Joshua Wendland, Markel Zubia, Roman Andriushchenko, Maris F. L. Galesloot, Milan Ceska, Henrik von Kleist, Thiago D. Simão, Maximilian Weininger, Nils Jansen)
rfPG: Robust Finite-Memory Policy Gradients for Hidden-Model POMDPs (Maris F. L. Galesloot, Roman Andriushchenko, Milan Ceska, Sebastian Junges, Nils Jansen)
A special highlight was the moment right after presenting On Evaluating Policies for Robust POMDPs, when we learned that the paper had just been accepted to NeurIPS 2025!
