Jakob Zech
Room 1/207
Im Neuenheimer Feld 205
69120 Heidelberg, Germany

jakob.zech@uni-heidelberg.de
+49 6221 5414504

Open PhD and Postdoc positions available, see here.

About

I am a Professor for the Mathematical Foundations of Machine Learning at the Institute for Mathematics at Heidelberg University. My background is in numerical analysis and approximation theory. Moreover I am affiliated with the Interdisciplinary Center for Scientific Computing. Before moving to Heidelberg in April 2020, I was a postdoc at MIT in the group of Youssef Marzouk. In 2018 I completed my PhD at ETH Zürich with Christoph Schwab.

Research

My research group investigates the mathematical foundations of Scientific Machine Learning (SciML). We combine rigorous analysis, probability theory, and deep learning to develop efficient, provably convergent algorithms for high-dimensional problems. Key focus areas include Operator Learning (neural operators for PDEs), Bayesian Inverse Problems, Structure-Preserving ML, Transport methods and Particle-based Optimization/Sampling.

Short CV

Since 9/2025	Professor at Heidelberg University
4/2020 - 8/2025	Assistant Professor at Heidelberg University
4/2019 - 3/2020	Postdoctoral fellow at MIT (SNSF fellowship 184530)
7/2014 - 3/2019	PhD in Mathematics at ETH Zürich
9/2012 - 6/2014	MSc in Applied Mathematics at ETH Zürich
10/2008 - 3/2012	BSc in Technical Mathematics at TU Wien