Machine learning in quantum physics
Machine learning and artificial intelligence are becoming important tools in many areas. We are interested in using machine learning in quantum physics, e.g., for aiding the preparation, detection or interpretation of quantum phases. Recently, we have employed artificial neural networks to identify topological phase transitions from experimental images and reconstructed the entire phase diagram of the topological Haldane model, despite a preparation at finite temperature. Next to supervised learning, we also used different unsupervised machine learning techniques in collaboration with the group of Maciej Lewenstein and Alexandre Dauphin at ICFO to obtain the phase diagram without any prior information. Such approaches will be crucial for more exotic phase diagrams and for potentially discovering unknown phases, which might not even have a conventional order parameter. We are also interested in the efficient representation of quantum states via neural networks and their applications in fast state tomography.
We offer theoretical Bachelor and Master projects in the area of machine learning in quantum physics, which provide a timely complement of a physics education. For further applications of machine learning within the CUI cluster, see the CUI task force machine learning.
We have recently organized an international conference on "Machine Learning in Natural Sciences: From Quantum Physics to Nanoscience and Structural Biology" in Hamburg. Check out the conference website for the recorded videos.
The project is funded by the Cluster of Excellence CUI and the DFG Research Unit FOR 2414.
recommended reading
- Benno S. Rem, Niklas Käming, Matthias Tarnowski, Luca Asteria, Nick Fläschner, Christoph Becker, Klaus Sengstock, Christof Weitenberg, Identifying quantum phase transitions using artificial neural networks on experimental data, Nature Physics 15, 917-922 (2019).
- Niklas Käming, Anna Dawid, Korbinian Kottmann, Maciej Lewenstein, Klaus Sengstock, Alexandre Dauphin, Christof Weitenberg, Unsupervised machine learning of topological phase transitions from experimental data, Machine Learning: Science and Technology 2, 035037 (2021).
Master and Bachelor theses finished within the team
- Moritz Sträter, Calculating Atom Number Fluctuations in a BEC and Concept for a Matter Wave Neural Network, Masterarbeit 2023
- Felix Herbort, KI-gestützte Auswertung und Optimierung von Quantengasexperimenten, Bachelorarbeit 2021
- Moritz Sträter, KI-gestützte Identifikation des BEC-Phasenübergangs, Bachelorarbeit 2021
- Tom Blöcker, GPU-accelerated time evolution for Gross-Pitaevskii equation solutions, Bachelorarbeit 2020
- Abdul-Rahman Rasul, Neuronale Netze für hochauflösende Messungen von Laserfrequenzen mittels Specklemustern, Bachelorarbeit 2020
- Bastian Lunow, Applications of Neural Networks in Quantum Gas Experiments, Masterarbeit 2020
- Niklas Käming, Analyzing Many Body Physics with Machine Learning, Masterarbeit 2020