SAMOP 2023 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
SYML: Symposium Machine Learning in Atomic and Molecular Physics
SYML 1: Machine Learning in Atomic and Molecular Physics
SYML 1.1: Invited Talk
Tuesday, March 7, 2023, 11:00–11:30, E415
An unsupervised deep learning algorithm for single-site reconstruction in quantum gas microscopes — •Alexander Impertro1,2,3, Julian F. Wienand1,2,3, Sophie Häfele1,2,3, Hendrik von Raven1,2,3, Scott Hubele1,2,3, Till Klostermann1,2,3, Cesar R. Cabrera1,2,3, Immanuel Bloch1,2,3, and Monika Aidelsburger1,2 — 1Department of Physics, Ludwig-Maximilians-Universität München, Schellingstr. 4, D-80799 Munich, Germany — 2Munich Center for Quantum Science and Technology (MCQST), Schellingstr. 4, D-80799 Munich, Germany — 3Max-Planck-Institut für Quantenoptik, Hans-Kopfermann-Strasse 1, D-85748 Garching, Germany
In quantum gas microscopy experiments, reconstructing the site-resolved lattice occupation with high fidelity is essential for the accurate extraction of physical observables. For short interatomic separations and limited signal-to-noise ratio, this task becomes increasingly challenging. Common methods rapidly decline in performance as the lattice spacing is decreased below half the imaging resolution. Here, we present a novel algorithm based on deep convolutional neural networks to reconstruct the site-resolved lattice occupation with high fidelity. The algorithm can be directly trained in an unsupervised fashion with experimental fluorescence images and allows for a fast reconstruction of large images containing several thousand lattice sites. We benchmark its performance using a quantum gas microscope with cesium atoms that utilizes short-spaced optical lattices with lattice constant 383.5nm and a typical Rayleigh resolution of 850nm. We obtain promising reconstruction fidelities around 96% across all fillings based on a statistical analysis. We anticipate this algorithm to enable novel experiments with shorter lattice spacing, boost the readout fidelity and speed of lower-resolution imaging systems, and furthermore find application in related experiments such as trapped ions.