///PhD Defence Henrik Riedel: „Optimizing CNNs in Civil Engineering“
Henrik Riedel has successfully defended his dissertation at TU Darmstadt. His research addresses key challenges in applying Convolutional Neural Networks (CNNs) to civil engineering, including limited data availability, high labeling costs, and model explainability.
A central outcome of his work is the Maximum Receptive Field (MRF) rule, which enables targeted optimization of CNNs and reduces the hyperparameter space. In addition, the dissertation shows that Explainable AI (XAI) methods can enhance trust in AI models while reducing computational effort.
These findings contribute to more efficient use of AI in the construction sector and provide valuable insights for the application of machine learning in engineering disciplines.
facadeworld 