A new Machine Learning result in Quantum Physics

John Bastian and Anton van den Hengel are among the authors of a new paper just published in Nature Scientific Reports.

The paper describes a Machine Learning-based approach to generating Bose-Einstein Condensates (BECs).  One of the difficulties in generating BECs is the need to cool Rubidium ions down to 10^{^-9} degrees Kelvin.  This is achieved partly using finely tuned lasers, and the Machine Learning process we developed was in charge of this part of the cooling.

BECs are groups of atoms that are so cold that they begin to behave as one.  BECs are thus used for very accurate measurement of magnetic and gravitational fields, amongst other things.  Measurements of these fields are particularly important for detecting new mineral deposits, and for precise navigation, but precise measurement of magnetic fields could also have important medical applications.

The Machine Learning technique used is a Gaussian Process, which has the advantage of providing a confidence measure associated with each estimate.  More details are available here.

The citation is Fast machine-learning online optimization of ultra-cold-atom experiments, P. B. Wigley, P. J. Everitt, A. van den Hengel, J. W. Bastian, M. A. Sooriyabandara, G. D. McDonald, K. S. Hardman, C. D. Quinlivan, P. Manju, C. C. N. Kuhn, I. R. Petersen, A. N. Luiten, J. J. Hope, N. P. Robins & M. R. Hush, Scientific Reports 6, Article number: 25890 (2016)