Question: what part of science are you best at

Harrison Prosper answered on 11 Jun 2019: last edited 11 Jun 2019 1:44 pm

I think I’m best at using probability theory and advanced methods, such as those based on machine learning, to extract signals from data. For example, from 1996 to 2009, I worked with a team of scientists from around the world to extract a signal (the creation of a top quark plus other particles that are not themselves top quarks) from data we collected at an accelerator called the Tevatron at Fermilab, near Chicago. I used a method called Bayesian neural networks to isolate the signal from the data and also developed a way to analyze the data, which, again for the first time in my field, correctly accounted for all kinds of complicated uncertainties. Some of these methods have been reused at the Large Hadron Collider at CERN, in particular, to search for particles that are predicted by theories based on a concept called supersymmetry. Supersymmetry can be thought of as the idea that there is no real distinction between matter and forces.