Research

As part of my research, I deal with the question of how distributed control systems can be made resilient to communication failures. The focus is on Decentralized Partially Observable Markov Decision Processes (Dec-POMDPs) and reinforcement learning algorithms.

Keywords:

• Reinforcement Learning
• Multi-agent systems
• Decentralized Control Systems
• Dec-POMDPs

Topics for theses

Feel free to contact me regarding a thesis on one of the above mentioned topics of my research. Furthermore, I offer theses in the following topics:

• Internet of Things
• 802.15.4-based protocols
• Graphics programming

---

Projects

• BMWK Project: RIWWER

---

Publications (3)

2026 (2)

• Zdankin, Peter and Kummerow, Arne and Weis, Torben
  CluBS: Clustering Behavioural Similarity
  In: 32nd SIGKDD Conference on Knowledge Discovery and Data Mining, 2026
• Josten, Malte and Kämmerer, Gérald and Kummerow, Arne and Weis, Torben
  Size Does Matter: The Impact of Embedding Models and Sizes on Spam Email Classification
  In: Proceedings of the 23rd International Conference on Security and Cryptography - Volume 1: SECRYPT
  DOI: 10.5220/0015082300004103

2023 (1)

• Kummerow, Arne and Weis, Torben
  Towards Distributed Control under Deficient Communication with Multi-agent Reinforcement Learning
  In: AUTOQUITOUS 2023: FIRST WORKSHOP ON AUTONOMOUS UBIQUITOUS SYSTEMS

---

Teaching

WiSe 26/27

• Rechnernetze und Kommunikationssysteme