I am a Postdoc at TU Munich in Massimo Fornasier's Applied Numerical Analysis group and a junior member of the Munich Center for Machine Learning. I completed my PhD in Applied Mathematics at TU Berlin under the supervision of Prof. Gabriele Steidl and as part of the Berlin Mathematical School.
My interests lie in optimal transport, primarily in Wasserstein gradient flows, and in kernel methods.
Here you will find the code for the following projects.
- Wasserstein Gradient Flows of Moreau Envelopes of f-Divergences in Reproducing Kernel Hilbert Spaces
- Wasserstein Gradient Flows of MMD Functionals with Distance Kernel and Cauchy Problems on Quantile Functions (see also this follow-up paper)
- Accelerated Stein variational gradient flow
- SympFormer: Accelerated attention blocks via Inertial Dynamics on Density Manifolds
viktor[dot]stein[at]tum.de
stein[at]math[dot]tu-berlin[dot]de