5 papers
Neural Scaling Laws for Learning-based Identification of Nonlinear Systems
Marco Roschkowski, Karim Cherifi, Hannes Gernandt
The use of machine learning models in system identification has increased due to their ability to approximate complex nonlinear dynamics with high accuracy. However, often it is no…
Long- and short-time behavior of hypocoercive evolution equations with higher index via modal decompositions
Marco Roschkowski, Hannes Gernandt
Hypocoercivity emerged in kinetic transport theory, allowing to derive exponential long-time estimates for evolution equations. Recently, the short-time asymptotics for equations w…
Two energy methods for distributed port-Hamiltonian systems and their application to stability analysis
Marco Roschkowski, Hannes Gernandt
We develop two local energy methods for distributed parameter port-Hamiltonian (pH) systems on one-dimensional spatial domains. The methods are applied to derive a characterization…
Improving Robustness of Foundation Models in Domain Adaptation with Soup-Adapters
Marco Roschkowski
In this paper, we tackle two fundamental problems in few-shot domain adaptation of foundation models. First, hyperparameter tuning is often impractical due to the lack of large val…
Nonlinear port-Hamiltonian system identification from input-state-output data (ISO-pHNN)
Karim Cherifi, Achraf El Messaoudi, Hannes Gernandt +1
In this paper, we introduce a framework called ISO-pHNN for identifying nonlinear port-Hamiltonian systems using input-state-output data. The framework utilizes neural networks' un…