Marten Lienen

Marten Lienen is a Research Fellow at the Technical University of Munich, affiliated with the TUM School of Computation, Information and Technology in the Department of Computer Science (I26 - Data Analytics and Machine Learning). He works within Prof. Stephan Günnemann's research group located at Boltzmannstr. 3 in Garching.

His research focuses on the intersection of generative modeling and physical systems simulation. Lienen specializes in developing machine learning approaches for time-dependent physical phenomena, particularly turbulent fluid flows. His work bridges deep learning with numerical methods for differential equations, creating novel frameworks that respect physical constraints while leveraging data-driven approaches.

Analyzing his publication record reveals a clear trajectory toward building physics-informed generative models capable of simulating complex dynamical systems. His most recent work on UnHiPPO demonstrates sophisticated integration of uncertainty quantification with state space modeling, while his publications on fluid dynamics simulation showcase practical applications of generative ML to longstanding challenges in computational physics.

Lienen maintains active online presences through his personal website (martenlienen.com), GitHub, and Twitter accounts, sharing research code and engaging with the machine learning community. His torchode library for PyTorch demonstrates commitment to creating practical tools for scientific machine learning.