Intelligent Systems

Bachelor and Master theses

the3

In the working group "Intelligent Systems" topics for theses projects are available at any time. In the following, we list some current examples of possible topics for bachelor and master theses - the topics are available in principle, but also offer a framework for what other topics might look like. Also, we list at the end of this page all the work already done.

In general, we try to define a topic together with the interested students. This means that you are welcome to contact us even with a rough idea or a direction in which you would like to work. We try to involve our students in current research projects so that the topics are of practical use for our research group.

Please note, however, that we assume that you already have all the necessary basic knowledge. This does not mean that you have to be an "expert" in the field of machine learning, for example - we are, however, not in a position to give you a basic course in the research and teaching area as part of a thesis project in addition to the learning content necessary for writing a student thesis. This means that ideally, you have already completed modules of our research group.

Besides, you should be able to work with the chosen programming language (i.e. primarily JAVA or Python) without any help. We also assume that you have a strong interest in machine learning and intelligent systems!

 

Open theses

You are also welcome to suggest your own topic for a thesis, please contact Prof. Tomforde

 


Ongoing theses

Bachelor

Is it possible for intersections in a simulated urban road network to establish "green waves" for traffic only with minimal information from the immediate neighbours? And how would this compare to an existing, more collaborative approach?
For more information: Thesis Green wave (pdf)

Previous theses

Bachelor

  • Erprobung von Reinforcement Lernverfahren in einer MAZE-basierten Laborumgebung
  • A Simulation-based Comparison of Incident Detection

Master

Investigation of possible improvements to increase the efficiency of the AlphaZero algorithm