I am a PhD student at Aarhus University, doing research in the intersection between machine learning and traditional simulation techniques. In recent years, neural networks (NNs) have been applied to a wide range of problems, from speech synthesis to image segmentation. More recently, approaches like neural ordinary differential equations (NODEs) and physics informed neural networks (PINNs) has started to infiltrate the modeling community, promising the ability to approximate complex non-linear dynamics.
In particular I am interested in how domain knowledge can be encoded as part of ML models to improve their accuracy and consistency with known physics. For ML my language of choice is Python combined with PyTorch, but for new research I prefer Jax. I also have an interest technologies like Functional Mock-up Interface (FMI) for which I have developed a tool for coupling models implemented in Python, C#, Java and MATLAB into existing simulation tools like OpenModelica and Dymola.
Outside of my research I enjoy traveling, spending time in the outdoors hiking, skiing and climbing - or on the ocean, scuba diving, spearfishing and sailing 🌄⛷🐟⛵🤿. Feel welcome to contact me if you are interested in collaborating or exchanging ideas.
🚧 Site currently under construction, content to be added soon™ 🚧