Andreas Bergmeister

I am a doctoral researcher in AI at TU Munich, in the group of Prof. Dr. Stefanie Jegelka and affiliated with the Munich Center for Machine Learning. I received my Bachelor's and Master's degrees in Computer Science from ETH Zurich and interned at the Swiss Data Science Center.

I am currently working on reinforcement learning for generative models, in particular diffusion models. Earlier, I worked on scaling graph generation, and developed PJAX, an alternative to autodiff frameworks for training neural networks without gradients.

Portrait of Andreas Bergmeister

Blog posts

Reinforce Adjoint Matching Scaling RL post-training of diffusion and flow-matching models. PJAX Training neural networks without backpropagation, using projection-based methods.

Selected publications