Viktor Stein

Postdoctoral researcher in applied mathematics working on Wasserstein gradient flows, regularized divergences, optimal transport, and geometry-inspired machine learning.

I am a postdoctoral researcher at the Chair of Applied Numerical Analysis in the Department of Mathematics at Technical University of Munich, led by Massimo Fornasier. I work on topics of the ERC Advanced Grant NEITALG and am a Junior Member of the Munich Center for Machine Learning (MCML).

I completed my PhD in June 2026 under the supervision of Gabriele Steidl in the Applied Mathematics Group at TU Berlin. During the final stage of my PhD, I was a Phase II student of the Berlin Mathematical School.

My research develops rigorous analytical foundations and practical algorithms for particle-based sampling methods, aiming to (i) understand properties of gradient flows in probability spaces and (ii) design fast, efficient, and stable algorithms.

The broader goal is to develop inference methods that are theoretically grounded, computationally efficient, and robust in modern machine learning applications.

Research interests

  • Metric gradient flows in Wasserstein geometry and kernelized variants
  • Kinetic extensions, including accelerated gradient flows
  • Particle methods for generative modeling
  • Design and analysis of loss functionals combining, e.g., optimal transport, $f$-divergences, and kernel distances
  • Infinite-dimensional geometry of probability measures