Accelerating linear algebra with graph neural networks
We’re pioneering the use of graph neural networks as a computational primitive for solving matrix problems.
Research
My overarching goal is to accelerate scientific and technological progress using machine learning and optimization. To exemplify what that can mean more concretely, I highlight a few examples of ongoing research projects below. Before moving to Uppsala University, I mainly worked on applications to medical imaging and radiotherapy planning. These are also described in more detail below.
For a complete list of publications, please refer to my Google Scholar.
Self-driving battery lab
We’re building an autonomous battery laboratory that combines robotics and AI into a closed-loop system for rapid experimentation and exploration.
Optimization for diffusion MRI
Our open-source library for numerical optimization of diffusion MRI experiments enables cutting-edge imaging technology on clinical scanners.
Automated radiotherapy planning
Radiotherapy planning is complex and time consuming. We’ve made dozens of inventions, primarily based on machine learning and optimization, that can guide or automate parts of the radiotherapy workflow.