AICON: Artificial Intelligence for Computer Networks

This project group is open to computer science and computer engineering students. It is intended to start in the summer term 2020.

In the next generation of networks (you may know it as “5G”), many network components will be moved to the cloud to improve the flexibility and manageability of our networks. In this scenario, called network function virtualization (NFV), virtualized network functions (VNFs), like firewalls or intrusion detection systems, which are installed in virtual machines or containers (Docker) and can be deployed on-demand on cloud infrastructures. These VNFs can then be connected (chained) to form more complex network services. This simplifies the development of novel network services, because deployments can be fully automated and be done in minutes instead of days, as required by legacy, hardware-based network functions.

Together with the new flexibility there are multiple challenges that arise, e.g., fine-grained control and coordination of where and when to start these VNFs in the cloud. Or how to schedule traffic flows between machines within a data center. Or how to optimize processing in a wireless scenario. In all these different aspects, AI/ML may help to automatically learn what to do and how to optimize networking - either completely by itself or supporting existing non-ML approaches.

In this project group, we will look into different aspects and challenges of modern computer networking and explore how AI/ML can be leveraged to solve these challenges. You will learn about latest research regarding AI/ML in networking, gain hands-on experience reproducing this research, and have room for your own ideas to design new and better AI/ML-driven approaches that go beyond existing state of the art. Finally, the outcomes of the project group should be documented and polished and ideally published in a scientific conference or journal.


  • Familiarize with latest AI/ML approaches in computer networking

  • Split into groups focusing on different aspects of computer networking

  • Implement and evaluate state-of-the-art AI/ML approaches from literature. Can you reproduce the reported results?

  • Do you see potential for improvement? Think of how to improve your implemented approach and come up with own ideas how to achieve even better results than the current state-of-the-art

  • Design, implement, and compare your own approach against the existing approaches from literature

  • Analyze and document all findings - and prepare for publication

  • Walk away with hands-on experience in two hot topics: AI/ML and networking!