An unsupervised deep learning algorithm for single-site reconstruction in quantum gas microscopes
Authors:
Alexander Impertro,
Julian F. Wienand,
Sophie Häfele,
Hendrik von Raven,
Scott Hubele,
Till Klostermann,
Cesar R. Cabrera,
Immanuel Bloch,
Monika Aidelsburger
Abstract:
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 resolu…
▽ More
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 $383.5\,$nm and a typical Rayleigh resolution of $850\,$nm. We obtain promising reconstruction fidelities~$\gtrsim 96\%$ 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.
△ Less
Submitted 22 December, 2022;
originally announced December 2022.
Discerning quantum memories based on electromagnetically-induced-transparency and Autler-Townes-splitting protocols
Authors:
Anindya Rastogi,
Erhan Saglamyurek,
Taras Hrushevskyi,
Scott Hubele,
Lindsay J. LeBlanc
Abstract:
Electromagnetically induced transparency (EIT) and Autler-Townes splitting (ATS) are similar,but different quantum optical phenomena: EIT results from a Fano interference, whereas ATS is described by the AC-Stark effect. Likewise, despite their close resemblance, light-storage techniques based on the EIT memory protocol and the recently-proposed ATS memory protocol (E. Saglamyurek et al. Nature Ph…
▽ More
Electromagnetically induced transparency (EIT) and Autler-Townes splitting (ATS) are similar,but different quantum optical phenomena: EIT results from a Fano interference, whereas ATS is described by the AC-Stark effect. Likewise, despite their close resemblance, light-storage techniques based on the EIT memory protocol and the recently-proposed ATS memory protocol (E. Saglamyurek et al. Nature Photonics 12, 2018) are distinct: the EIT protocol relies on adiabatic elimination of absorption, whereas the ATS protocol is based on absorption. In this article, we elaborate on the distinction between EIT and ATS memory protocols through numerical analysis and experimental demonstrations in a cold rubidium ensemble. We find that their storage characteristics manifest opposite limits of the light-matter interaction due to their inherent adiabatic vs. non-adiabatic nature. Furthermore, we determine optimal memory conditions for each protocol and analyze ambiguous regimes in the case of broadband storage, where non-optimal memory implementations can possess characteristics of both EIT and ATS protocols. We anticipate that this investigation will lead to deeper understanding and improved technical development of quantum memories, while clarifying distinctions between the EIT and ATS protocols.
△ Less
Submitted 7 February, 2019;
originally announced February 2019.
