1st International Workshop on Advanced Deep Reinforcement Learning in Energy Systems (ADELE '25)
Dr.-Ing. Eric Veith
Department of Computing Science (» Postal address)
Lectures
Summer term 2025
1st International Workshop on Advanced Deep Reinforcement Learning in Energy Systems (ADELE '25)
Scope
The 1st International Workshop on Advanced Deep Reinforcement Learning for Energy Systems (ADELE) welcomes submissions from academic and industry focusing on Deep Reinforcement Learning and its applications to the energy domain. The workshop is geared towards presenting, exploring, and discussing newest approaches from the DRL domain and inspiring quick transition towards application and further research by scientists in the energy domain. ADELE recognizes that the premier targets of DRL research is not the energy domain and, therefore, strives to foster close cooperation between DRL and energy scientists.
Topics of interest include, but are not limited to:
- Approaches to convering large state/action spaces, including handling underlying physical properties and controller conflicts
- State space representations
- Approachs to eXplainable Deep Reinforcement Learning suitable to large and complex (e.g., mixed discrete-continuous) state/action spaces
- Safe and Risk-averse DRL
- Offline learning from domain user knowledge without explicit trajectory encoding
- Physics-informed or model-based DRL
- Modifications of existing model-free algorithms wrt. the energy domain
- Neuroevolutionary DRL
- Hybrid agents (e.g., DRL-based extensions of known controllers)
- Agent verification
Workshop Schedule
| Time | Presentation |
|---|---|
| 09:00 – 09:30 | Workshop Keynote: Integrating Weather and Energy Systems in the Era of Weather Foundation Models Prof. Dan Wang, Hong Kong Polytechnic University |
| 09:30 – 10:00 | Reinforcement Learning with Partially Defined Rewards and Human Feedback for Energy Efficiency Recommendations S. Chadoulos, I. Koutsopoulos, G. Polyzos, N. Ipiotis |
| 10:00 – 10:30 | Improving Demand Response Programs Using Override Signals with Reinforcement Learning S. Kumar, A. Easwaran, B. Delinchant, R. Rigo-Mariani |
| 10:30 – 11:00 | Coffee Break |
| 11:00 – 11:30 | Efficient Multi-Objective Optimisation for Real-World Power Grid Topology Control Y. El Manyari, A. R. Fuxjäger, S. Zahlner, J. Van Dijk, A. Castagna, D. Barbieri, J. Viebahn, M. Wasserer |
| 11:30 – 12:00 | Multi-objective Reinforcement Learning for Smart Heating Controller (Poster Presentation) F. Carton |
| 12:00 – 12:30 | Cover Me: Mitigating Multi-Agent System Failure through Reinforcement Learning—A Technology Demonstration E. Veith, A. Wellßow, T. Logemann, E. Frost |
For all presentations, we encourage the speaker to stay within a 20-minutes timeframe to allow for discussions.
Submission
Fomat
ADELE invites submissions in the following format:
- Full workshop papers, up to 10 pages, following the ACM template according to the main conference’s guidelines (https://energy.acm.org/conferences/eenergy/2025/cfp.php).
- Tutorials, with an accompaning paper of up to 4 pages, following the same templates as a full workshop paper
Site
The submission site of ADELE is located at: https://adele25.hotcrp.com/
Submissions will be peer-reviewed by at least 3 reviewers. Presentations will honour the workshop format by providing 20 minutes with the actual slide deck and additional 10 minutes for discussion
Important Dates
| Submission Deadline | April 1st, 2025, 23:59 AoE |
| Notification | April 15th, 2025, 23:59 AoE |
| Camera Ready | May 1st, 2025, 23:59 AoE Final Version in ACM TAPS |
Programm Committee
Omid Ardakanian, University of Alberta, Canada
Dan Wang, The Hong Kong Polytechnic University, Hong Kong, China
Florence Carton, Total Energies, France
Tianyu Zhang, Autodesk Research, Canada
Eric MSP Veith, Carl von Ossietzky University Oldenburg
Arlena Wellßow, Carl von Ossietzky University Oldenburg
Torben Logemann, Carl von Ossietzky University Oldenburg