Michael Feischl

Michael Feischl is a Professor for Computational PDEs at TU Wien (since 2022) and holds an ERC Consolidator Grant for his project "New Frontiers in Optimal Adaptivity" (2024–2029). His research focuses on partial differential equations with random coefficients, computational micromagnetism (Landau-Lifshitz-Gilbert equation), and optimal adaptive mesh refinement techniques. He leads the Computational PDEs research group within the Institute of Analysis and Scientific Computing.

Education and career highlights include roles as Associate Professor at TU Wien (2019–2022), W2 Professor at University of Bonn (2017–2018), and Junior Research Group Leader at KIT (2015–2017). His work bridges numerical analysis, computational physics, and machine learning, with a strong emphasis on rigorous mathematical foundations and algorithmic efficiency.

Research interests include:

• Adaptive finite element and boundary element methods
• Stochastic modeling and uncertainty quantification
• Computational methods for micromagnetic simulations
• Machine learning applications in numerical analysis

His recent work explores optimal adaptivity for time-dependent PDEs, neural network-based solvers, and efficient discretization strategies for complex physical systems. Key contributions include advancements in a posteriori error estimation and hierarchical training of neural networks.