The Third International Workshop on Machine Learning in Networking (MaLeNe, https://www.uni-augsburg.de/en/fakultaet/fai/informatik/prof/netcom/events/malene2025/) was a successful half day event held at Technical University of Ilmenau on September 1, 2025, where it was co-located with the Conference on Networked Systems (NetSys 2025). It follows the history of MaLeNe KuVS workshops (Fachgespräch) and MaLeNe NetSys workshops as the 6th edition of the MaLeNe series, and was organized by workshop co-chairs Michael Seufert (University of Augsburg, Germany), Andreas Blenk (Siemens AG, Germany), and Björn Richerzhagen (Siemens AG, Germany). The workshop accepted 7 full papers for presentation.

On the day of the workshop, the co-chairs welcomed 34 registered participants. The workshop started with an industry keynote given by Philippe Buschmann (Siemens AG, Germany) who elaborated on “Recent AI Trends in Industrial Network Environments”. He explored industrial AI and its application in smart mobility, smart infrastructure, and manufacturing. He especially highlighted the challenges and opportunities of AI adoption in these settings through the perspective of Siemens.

Afterwards, the first technical session started. Katharina Dietz (University of Würzburg, Germany) evaluated the consensus of XAI -based network intrusion detection and how to improve consensus with feature selection methods. Nasim Nezhadsistani (University of Zurich, Switzerland) presented an asynchronous consensus-driven multi-agent approach to decentralized federated learning for intrusion detection in 5G networks. Zineddine Bettouche (Deggendorf Institute of Technology, Germany) discussed spatiotemporal machine learning techniques for cellular traffic forecasting with a particular focus on Mamba deep learning architectures.

After the coffee break, Christian Maier (Salzburg Research, Austria) elaborated on predicting performance metrics in edge-cloud networks using GNNs. Timothy Harrison (University of Hagen, Germany) presented work on state cloning for high order Markov chains with the GraphLearner. Yanakorn Ruamsuk (University of Hagen, Germany) talked about emotion-controlled communication in agent networks. Finally, Alexander Niedermayer (Karlsruhe University of Applied Sciences, Germany) presented an approach for client-agnostic continuous authentication via keystroke-induced traffic patterns.

The workshop co-chairs closed the day with a short recap and thanked all speakers and participants who engaged in the fruitful discussions. As the workshop has proven to foster active collaborations in the research community, another edition will be considered in the future.

We would like to thank all the authors, reviewers, and attendants for their precious contributions towards the successful organization of the workshop!

Michael Seufert, Andreas Blenk, Björn Richerzhagen
MaLeNe 2025 Workshop Co-Chairs