Berlin 2024 – wissenschaftliches Programm

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

DY: Fachverband Dynamik und Statistische Physik

DY 34: Poster: Machine Learning, Data Science, and Reservoir Computing

DY 34.2: Poster

Mittwoch, 20. März 2024, 15:00–18:00, Poster C

Optical Ising model simulations with caesium vapor cells — •Kilian Junicke¹, Elizabeth Robertson^1,2, Mingwei Yang^1,2, Inna Kwiatkowski^2,3, and Janik Wolters^1,2 — ¹Technische Universiät Berlin, Institute for Optics and Atomic Physics, Hardenbergstr. 36, 10623 Berlin, Germany — ²Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR), Rutherfordstr. 2, 12489 Berlin, Germany — ³TU Berlin, Institut für Luft und Raumfahrt

Several computationally hard optimization problems can be mapped to finding the ground state of an Ising model [1]. Simulating Ising models optically promises speed increases [2]. Building an optical Ising machine then raises the question of how to simulate the spin states [3].

Here we present a scheme for simulating an Ising model using the ground states of cesium vapor at room temperature. We present methods for implementing positive and negative interactions using a measurement and feedback strategy. In the system electromagnetically induced transparency acts as a frequency transducer. We initialize the system and allow it to evolve by executing a series of pump probe operations on spatially multiplexed regions of an atomic vapor cell until a ground state solution is found.

[1] Lucas, A. Ising formulations of NP problems. Front. Phys. 2, 5 (2014).

[2] McMahon, P.L. Physics of optical computing. Nat Rev Phys 5, 717-734 (2023).

[3] Böhm et al. Poor man's coherent Ising machine for optimization. Nat Commun 10, 3538 (2019).

Keywords: Ising Model; Electromagnetically induced transparency; atomic vapor; optical pumping