Hannover 2020 – scientific programme
The DPG Spring Meeting in Hannover had to be cancelled! Read more ...
Parts | Days | Selection | Search | Updates | Downloads | Help
Q: Fachverband Quantenoptik und Photonik
Q 4: Quantum gases (Fermions) I
Q 4.7: Talk
Monday, March 9, 2020, 12:30–12:45, e214
Unsupervised Machine Learning of Topological Phase Transitions from Experimental Data — •Niklas Käming1, Benno Rem1,2, Matthias Tarnowski1,2, Luca Asteria1, Nick Fläschner1, Christoph Becker1,3, Klaus Sengstock1,2,3, and Christof Weitenberg1,2 — 1ILP - Institut für Laserphysik, Uni- versität Hamburg, Luruper Chaussee 149, 22761 Hamburg, Germany — 2The Hamburg Centre for Ultrafast Imaging, Luruper Chaussee 149, 22761 Hamburg, Germany — 3ZOQ - Zentrum für Optische Quantentechnologien, Universität Hambur g, Luruper Chaussee 149, 22761 Hamburg, Germany
Machine learning techniques such as artificial neural networks are currently revolutionizing many technological areas and have also proven successful in quantum physics applications. Here we apply unsupervised learning techniques such as deep convolutional autoencoders to identify phase transition from experimental data. We map out the complete two-dimensional topological phase diagram of the Haldane model with an unsupervised learning scheme applied to experimental momentum space density images of ultracold quantum gases. Our work points the way to experimentally explore unknown complex phase diagrams without prior knowledge of the underlying theoretical structure.