RESEARCH

We study models for quantum information processing and fundamental aspects of quantum information theory. One focus of our research is the theory of measurement-based quantum computation, which has resulted in a new and more thorough understanding of many-body entanglement as resource, and applications in quantum communication, quantum error correction, and quantum algorithms. Another focus lies on the application and development of machine learning in basic science and on the study of physical models for learning. Some of our work is highly interdisciplinary and addresses questions in different fields including quantum physics, robotics, behavioural biology, and the philosophy of action.

AI and Science

Quantum Computation

... and more

News & Activities

Paulin de Schoulepnikoff, PhD studen in the group, has won the poster prize for his work on interperable machine learning for quantum phenomena both at the SFB BeyondC Conference 2026 and the UIBK Day of Scientific Computing 2026. Congratulations Paulin!

New publication: Information Thermodynamics of Agents: The Work Capacity of Channels with Memory
In our new PRX paper, we show that agents must forget parts of their past actions to operate at optimal thermodynamic efficiency, pointing toward new design principles for energy-efficient artificial agents. Check out the news coverage in Phys.org and Der Standard!

Joséphine Pazem, PhD student in the group, has won the poster prize for her work on finding canonical causal representations of a system to identify agent systems from actions at the IWAI 2026 conference. Congratulations Joséphine!
Josephine poster prize

New preprint: Emergent Aggregation from Collective Foraging
Our new work on foraging, this time investigating a collection of agents observing and interacting with each other, just hit the arxiv! This is a collaboration with the Soft Matter Lab group at the Gothenburg University.

New preprint: Representational separation between unitary and channel quantum generative models via shared classical randomness at shallow depth
We show that shared classical randomness, a comparatively weak resource from entanglement theory, is sufficient to establish such a strict scalable representational separation over the corresponding shallow unitary Born model.