Jonas Kusch

Jonas Kusch is an Associate Professor at the Department of Data Science, Norwegian University of Life Sciences, specializing in numerical analysis and its applications in scientific computing and machine learning. His work focuses on dynamical low-rank approximation, particularly in developing low-rank neural networks with geometry-aware training algorithms that respect the differential geometry of matrix manifolds.

• Research spans computational quantum mechanics, radiation transport, and machine learning.
• Key contributions include energy-stable integrators for kinetic equations and multi-fidelity optimization algorithms for fission criticality.
• Publications emphasize low-rank methods for solving inverse problems, time-dependent systems, and uncertainty quantification in hyperbolic equations.