On May 13, 2026, our colleague Max Zimmer ((AISST/IOL) successfully defended his doctoral dissertation titled “Effective and Efficient Prune-Retrain Pipelines for Neural Network Compression” at the Technische Universität Berlin.
In his research, Max Zimmer explores innovative methods for compressing neural networks—an important topic in the field of machine learning. The goal is to significantly improve the efficiency of high-performing models so they can be deployed on resource-constrained systems. The dissertation focuses on so-called prune-retrain pipelines, which systematically remove redundant model parameters and subsequently fine-tune the resulting models without incurring significant performance loss.
With his work, Max Zimmer makes an important contribution to the advancement of efficient AI systems and the practical applicability of modern machine learning methods.
We warmly congratulate him on his successful defense!
