My research agenda is focusing on the intersection between Deep Learning, the physical sciences, and scientific computing. Current projects include:

• Generative AI Based Optimization: conceptually new approaches to Combinatorial Optimization problems based on Generative Deep Learning and Statistical Mechanics.
• Data-efficient Deep Learning: design of mathematically founded training strategies for Deep Neural Networks in scenarios with little training data from physical systems.
• Learning from Quantum Teachers: exploring the potential of Deep Learning using data from quantum computers.

• L. Lewis, H. Huang, V. Tran, S. Lehner, R. Kueng, J. Preskill, Improved machine learning algorithm for predicting ground state properties, Nature Communications 15(1): 895, 2024
   
• S. Sanokowski, W. Berghammer, S. Hochreiter, S. Lehner, Variational Annealing on Graphs for Combinatorial Optimization, Neural Information Processing Systems (NeurIPS 2023)
   
• A. Mayr, S. Lehner, A. Mayrhofer, C. Kloss, S. Hochreiter, J. Brandstetter, Boundary graph neural networks for 3d simulations, Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI 2023)
   
• P. M. Winter, C. Burger, S. Lehner, J. Kofler, T. I. Maindl, C. M. Schäfer, Residual neural networks for the prediction of planetary collision outcomes, Monthly Notices of the Royal Astronomical Society 520 (1), 1224-1242, 2023
   
• M. Gauch, M. Beck, T. Adler, D. Kotsur, S. Fiel, H. Eghbal-zadeh, J. Brandstetter, J. Kofler, M. Holzleitner, W. Zellinger, D. Klotz, S. Hochreiter, S. Lehner, Few-Shot Learning by Dimensionality Reduction in Gradient Space, Proceedings of the 1st Conference on Lifelong Learning Agents (CoLLAs 2022)

Full list: Google Scholar profile, opens an external URL in a new window

My teaching activities involve a wide range of classes in the AI curricula, ranging from Introductory AI lectures to Master’s Thesis seminars. For an exhaustive list please follow this link, opens an external URL in a new window and select the year and term on the top right.