DPG Phi
Verhandlungen
Verhandlungen
DPG

SAMOP 2023 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

SYML: Symposium Machine Learning in Atomic and Molecular Physics

SYML 1: Machine Learning in Atomic and Molecular Physics

SYML 1.2: Hauptvortrag

Dienstag, 7. März 2023, 11:30–12:00, E415

Physics-inspired learning algorithms for optimal shaping of atoms with light — •Maximilian Prüfer — Vienna Center for Quantum Science and Technology, Atominstitut, TU Wien

Nowadays the high degree of control over optical potentials is key to many quantum simulations performed with ultracold atomic systems. In this talk I will show how arbitrary optical potentials can be created using, e.g., digital micromirror devices. Experimentally it is advantageous to optimize the desired potentials 'offline', that is not using the actual experiment but a digital twin trained using machine learning methods. In our new approach we use a physics-inspired model with few parameters combined with an iterative algorithm based on Iterative Learning Control. These methods allow for model-based 'offline' optimization as well as experimental feedback-based 'online' optimization which leads to an order of magnitude faster optimization compared to heuristic methods.

100% | Mobil-Ansicht | English Version | Kontakt/Impressum/Datenschutz
DPG-Physik > DPG-Verhandlungen > 2023 > SAMOP