Workshop on Reinforcement Learning 2025
Welcome to the second mini-Workshop on Reinforcement Learning (RL) in Mannheim. The workshop will be held on February 7th 2025 to gather enthusiasts in the field of theory of Reinforcement Learning (RL). In a relaxed athmosphere experts and beginners have the opportunity to exchange insights on current research on RL.
We extend a warm invitation to all interested individuals to join us for this collaborative event. Please note that there are no fees associated with the workshop, so just drop by if you are interested in reinforcement learning!
Speakers and titles
- Peter Dayan (Tübingen): Risk sensitive planning and exploration
- Daniele Calandriello (Deepmind): TBA
- Alberto Maria Metelli (Milano): Importance Weighting in Policy Gradient Methods
- Michael Muehlebach (Tübingen): On the Sample Complexity of Online Reinforcement Learning: A Multi-Model Perspective
- Gergely Neu (Barcelona): Optimal transport distances for Markov chains
- Alizée Pace (Zurich): Preference Elicitation for Offline Reinforcement Learning
- Davide Tateo (Darmstadt): Safe and Robust Real-World Robot Reinforcement Learning
Program
9:00–9:45 Calandriello 9:45–10:30 Dayan cofee break + poster session 11:15–12:00 Muehlebach lunch + poster session 14:00–14:45 Tateo 14:45–15:30 Pace coffee break + poster session 16:00–16:45 Neu 16:45–17:30 Metelli restaurant Postersession
To begin the discussions, there will be poster sessions during the breaks. Everyone is welcome to bring posters from other occasions as well.
Venue
University of Mannheim (Germany), Barockschloss, Gartensaal (here is a google maps link, almost exactly where the red dot is). Or check out the header of this website, just in the middle of the picture, the middle massive door.
Registration
We do not charge a fee for participation but ask participants to register for a better planning of coffee breakes and lunch. Please use our google forms sheet.
Organizers
Leif Döring (Mannheim), Claire Vernade (Tübingen), Simon Weißmann (Mannheim)
Support
The workshop is supported by the Mannheim Center for Data Science and the Deutsche Forschungsgemeinschaft (DFG) through the Emmy Noether project FoLiReL VE 1412/
1–1.