Alexander Hinterleitner, a doctoral candidate at the institute, presented the paper „Tuning for Explainability: Incorporating XAI Consistency into Multi-Objective Hyperparameter Optimization“. This work was conducted in collaboration with industry partner Everllence SE. Additionally, Aleksandr Subbotin presented the paper „Physics-Informed Neural Networks for State Estimation Problem“. This study proposes a method integrating Physics-Informed Neural Networks (PINNs) with the Unscented Kalman Filter (UKF) to address state estimation in nonlinear dynamical systems. Here are some impressions from their talks.
The proceedings are Open Access and can be downloaded here: https://publikationen.bibliothek.kit.edu/1000186052
