Tracking normal fluid flow in He II with unsupervised machine learning

Time dependent observations of fluid flow around large objects in three dimensions under extreme conditions are necessary to measure point-to-point correlations of the velocity vector field (structure functions) resulting from flow perturbed by objects. Using thermal gradients, we induced flow of the normal fluid component of liquid He II and observed the flow by recording fluorescence of  excimers produced by neutron capture throughout a ~cm³ volume. We applied an unsupervised machine learning algorithm to identify individual excimer clouds and then track their motion with millimeter and millisecond precision. Owing to the fact excimers are produced over a large region, the data are sparse in comparison to other techniques to produce excimers.  Machine learning is crucially important to track flow represented by sparse data and its importance will increase as improvements are made to overcome the sparsity of data.