Summary of the Dissertation

Reinforcement learning (RL) is a promising method of machine learning in which an agent learns control strategies solely through interaction with its environment.

As part of this doctoral thesis, the use of RL in large-scale industrial plants with high repetition rates was investigated. To demonstrate this concept, the KINGFISHER system—based on the novel AMD/Xilinx Versal platform—was designed and implemented to support experimental applications.

It was deployed at the KARA synchrotron at KIT, where it successfully learned to control beam instabilities without any prior knowledge—making it the first system trained entirely online on hardware at a particle accelerator.

We congratulate Luca Scomparin on this outstanding achievement!