Successful promotion of Kevin Kollek
On 28 March 2025, Mr Kevin Kollek successfully completed his doctorate at the Faculty of Electrical Engineering, Information Technology and Media Technology at the University of Wuppertal.
In his dissertation entitled "Contributions to the optimisation of artificial neural networks - compression methods for artificial neural networks, realisation of efficient algorithms on embedded systems and efficient approximation of model performance" (Shaker Verlag, Aachen 2025, ISBN 978-3-8191-0174-8), Mr Kollek deals with the question of how artificial neural networks can be optimised for use on resource-constrained embedded platforms without losing accuracy or robustness.
The work focuses on methods for compressing and accelerating neural networks, which significantly reduce memory requirements and computational complexity. In addition, Mr Kollek is developing implementation strategies for efficient deep learning algorithms on embedded processor architectures in order to be able to operate complex models even under real-time requirements. In addition, he presents methods for approximating model performance, which enable an early assessment of the performance of alternative network architectures.
The dissertation thus makes a significant contribution to the connection between artificial intelligence and embedded systems, a field of research that is becoming increasingly relevant, particularly for energy-efficient signal and image processing in modern IoT, industrial and mobility applications.
Prof Dr Anton Kummert will be the keynote speaker and Prof Dr Anselm Haselhoff (Westfälische Hochschule Gelsenkirchen) will be the co-lecturer.
The Chair of Theoretical Communications Engineering and Digital Signal Processing congratulates Dr Kevin Kollek on successfully completing his doctorate and wishes him all the best and every success for his future scientific and professional career.
