Call For Papers

The Third International Workshop on Machine Learning in Networking (MaLeNe) will be held in conjunction with the Conference on Networked Systems (NetSys 2025) in Ilmenau, Germany, on September 1, 2025.

In recent years, communication networks have become highly flexible through the employment of virtualization and softwarization paradigms. Still networks are highly complex, dynamic and time-varying systems, such that the statistical properties of networks and network traffic cannot be easily understood and modeled. Furthermore, the interplays between networking and the dynamic and heterogeneous requirements, expectations, and experiences of applications and users are increasing the complexity of the systems, which makes fault, configuration, performance, and security management in networks a hard problem. As observed in other disciplines, the successful application of machine learning can help to overcome these issues by following a more data-driven approach. Also in the networking domain, the technological advancements in the area of machine learning, the increasing availability of network analytics data, and the flexibility of programmable networks and virtualized network resources have made this approach applicable, which creates exciting new opportunities. 

MaLeNe 2025 aims at providing an international forum for researchers addressing emerging concepts and challenges related to machine learning in networking. The workshop will aim to address opportunities where machine learning can bring benefits to networking in different facets, such as network monitoring, management, and security. Together with flexible and programmable networks this paves the way towards a more proactive and autonomous network design and “self-driving” networks. The long-term vision is that configuration decisions can be made in real-time in an automated fashion before service and experience degradation occurs. The workshop will combine original paper presentations with a motivating keynote to thoroughly explore this challenging topic.

Topics of Interest

Authors are invited to submit papers that fall into or are related to the topic areas listed below:

• Methodology
  • Data sets for benchmarking, verification, proof of concept
  • Data augmentation
  • Performance evaluation methodology (best practices)
  • Good standards for data publishing
  • Data prediction and generation (e.g., GANs)
  • Dimensionality reduction (e.g., autoencoder)
• Artificial Intelligence and Machine Learning Methods
  • Classical methods like supervised, unsupervised, reinforcement learning
  • Deep methods vs non-deep methods
  • Advanced methods like adversarial, transfer
  • Large Language Models
  • TinyML
• Generalizability
  • Transfer of trained models (e.g., small to large networks, enterprise to data center)
  • Federated learning (combine models trained for different data sets)
  • Machine unlearning
  • Catastrophic forgetting
• Explainability
  • Explainable Artificial Intelligence (XAI)
  • Visualization
  • Understanding decisions of ML-based systems (management, traffic engineering, etc.)
  • Game-theory-based approaches to approximate guarantees
• Networking for Machine Learning and AI
  • Network architectures
  • Network applications
  • Network use cases (data center, enterprise, etc.)
  • Network resource management (algorithms, schedulers etc.)
  • In-network processing
• Applications in Networking
  • Network monitoring, especially from encrypted traffic (e.g., traffic classification, QoE)
  • Network configuration (e.g., suggest optimal configurations, “spell-check” text-based configuration data)
  • Network planning (e.g., reconfigurable data centers, job placement)
  • Network management (e.g., autonomous management, self-driving networks)
  • Network security (e.g., intrusion detection, covert channels, firewall)
  • Advanced networks (e.g., 5G to 6G, industry, slicing)
• Hot Topics from Machine Learning
  • Self-supervised learning
  • Intrinsic motivation, empowerment, curiosity
  • Language processing in networking
  • Meta-artificial intelligence (learning to learn)
  • Agentic AI
  • Foundation Models
  • From Large to Small language models

Paper Submission

Paper submissions must present original, research or experiences. Late-breaking advances and work-in-progress reports from ongoing research are also encouraged. Only original papers that have not been published or submitted for publication elsewhere can be submitted. Also extended versions of conference or workshop papers that are already published may be considered as long as the additional contribution is at least 30% new content from the original. Each submission must be written in English, accompanied by a 75 to 200 word abstract and a list of up to 5 key words. There is a length limitation of 6 A4 (210 mm x 297 mm) pages for full papers (including title, abstract, figures, tables) plus 1 page for references. Submissions must be in 2-column IEEE conference style with a minimum font size of 10 pt. Papers exceeding these limits, multiple submissions, and self-plagiarized papers will be rejected without further review. Authors should submit their papers electronically via the EasyChair online submission system under the following link: https://easychair.org/conferences/?conf=malene2025

Important Dates

• Paper submission deadline: June 6, 2025
• Paper notification: July 18, 2025
• Camera-ready deadline: July 25, 2025

Proceedings

Papers accepted for MaLeNe 2025 will be included in the conference proceedings or in separate workshop proceedings (such as OPUS). We reserve the right to remove any paper from the proceedings if the paper is not presented at the workshop.

Organizers

• Michael Seufert (University of Augsburg)
• Andreas Blenk (Siemens AG)
• Björn Richerzhagen (Siemens AG)