Greetings from the APS March Meeting in snowy Boston, Massachusetts. For me this first day has been about machine learning – specifically, the use of such techniques to analyze and interpret experimental data. It is very exciting to see a number of sessions at this conference devoted to that, though as a field it is only just about beginning to take shape. Do we want this to be about simplifying and speeding up experimental data analysis, thus helping experimentalists make contact with theory? Or is there potential for a deeper transformation, one where perhaps the meaning of the word “theory” will change, as machines take control of larger and larger chunks of the process we call Condensed Matter Physics? This morning I gave a talk by the above title where I hinted at that – and interesting discussions with my esteemed colleagues followed. Here’s a link to the abstract on the official meeting website.
A machine-learning approach to magnetic neutron scattering
