Non-destructive shadowgraph imaging of ultra-cold atoms.

An imaging system is presented that is capable of far-detuned non-destructive imaging of a Bose-Einstein condensate with the signal proportional to the second spatial derivative of the density. Whilst demonstrated with application to Rb85, the technique generalizes to other atomic species and is shown to be capable of a signal-to-noise of ∼25 at 1 GHz detuning with 100 in-trap images showing no observable heating or atom loss. The technique is also applied to the observation of individual trajectories of stochastic dynamics inaccessible to single shot imaging. Coupled with a fast optical phase locked loop, the system is capable of dynamically switching to resonant absorption imaging during the experiment.

Simultaneous Precision Gravimetry and Magnetic Gradiometry with a Bose-Einstein Condensate: A High Precision, Quantum Sensor.

A Bose-Einstein condensate is used as an atomic source for a high precision sensor. A 5×10^{6} atom F=1 spinor condensate of ^{87}Rb is released into free fall for up to 750 ms and probed with a T=130 ms Mach-Zehnder atom interferometer based on Bragg transitions. The Bragg interferometer simultaneously addresses the three magnetic states |m_{f}=1,0,-1⟩, facilitating a simultaneous measurement of the acceleration due to gravity with a 1000 run precision of Δg/g=1.45×10^{-9} and the magnetic field gradient to a precision of 120 pT/m.

Fast machine-learning online optimization of ultra-cold-atom experiments.

We apply an online optimization process based on machine learning to the production of Bose-Einstein condensates (BEC). BEC is typically created with an exponential evaporation ramp that is optimal for ergodic dynamics with two-body s-wave interactions and no other loss rates, but likely sub-optimal for real experiments. Through repeated machine-controlled scientific experimentation and observations our 'learner' discovers an optimal evaporation ramp for BEC production. In contrast to previous work, our learner uses a Gaussian process to develop a statistical model of the relationship between the parameters it controls and the quality of the BEC produced. We demonstrate that the Gaussian process machine learner is able to discover a ramp that produces high quality BECs in 10 times fewer iterations than a previously used online optimization technique. Furthermore, we show the internal model developed can be used to determine which parameters are essential in BEC creation and which are unimportant, providing insight into the optimization process of the system.