About me

I’m a PhD student and research associate at the University of Siegen at the groups for Computer Vision (lead by Prof. Michael Möller) and Visual Computing (lead by Prof. Jovita Lukasik). My research concerns the usage of machine learning approaches for inverse problems in imaging, with a focus on hybrid methods, combining the mathematical guarantess of classical, variational methods with the performance of modern, deep learning based solutions. Another focus is the use of methods of the field of neural architecture search to find promising solution candidates for these problems.

Teaching & Supervision

I supervise the bachelor courses “Digitale Bildverarbeitung” and “Digitale Bildverarbeitung - Praktikum”, and the master course “Unsupervised Deep Learning”. For everything related to these courses please refer to the Moodle-Pages or contact me via Mattermost (@alexander.auras1) or E-Mail.

For supervision of bachelor/master theses or Studienarbeiten please refer to the Websites of the Computer Vision group and the Visual Computing group.

News

• 16. Oct. 2025: This website is finally online.
• 13. Oct. 2025: I’ll supervise the exercises of the new Unsupervised Deep Learning lecture this semester.
• 1. Oct. 2025: I’ve joined the group for Visual Computing, lead by Prof. Jovita Lukasik.
• 25. Sep. 2025: We’ve submitted a paper to the ICLR 2026.

Recent Publications

• Aug. 2024: Robustness and exploration of variational and machine learning approaches to inverse problems: An overview
  Providing an overview of current approaches for solving inverse problems in imaging using variational methods and machine learning.
• Feb. 2024: Convergent Data-Driven Regularizations for CT Reconstruction
  Proving the convergence with regards to the noise level of two regularization methods for 2D parallel beam CT-reconstruction, and investigating the effect of discretization errors at different resolutions.