Hurray! The pivot article that marks my transition from physics to statistic modeling is finally out:

How to estimate the differential acceleration in a two-species atom interferometer to test the equivalence principle

G Varoquaux, R A Nyman, R Geiger, P Cheinet, A Landragin and P Bouyer

To put things in context, at the end of my PhD, we had been building an atom interferometer to test the Einstein equivalence principle and my reflections on the limits of atom interferometry shifted from worrying about the underlying physics, to worrying about the estimation: the inverse problem of going from the experimental signal, to the underlying quantities that we are measuring, confounded by all the horrible experimental noise.

The problem is: we want to do high precision metrologic tests in a free-falling plane. We use interferometry to measure gravity fields. But rather than doing interferometry with light, we use atoms, that are much more coupled to gravity. When probing gravity fields with light, the trick is to use hug