Professor Wasserman will discuss the application of statistics to a problem in particle physics. The goal is to be able to test for a certain event (the production of two Higgs bosons) in a set of data coming from a particle accelerator. To construct a hypothesis test, we need to know what the distribution of events would look like if there were no signal. We use a tool called "optimal transport'' to estimate this distribution. He will first give some background on optimal transport and then show how it can be used for this problem.
This is joint work with: Tudor Manole, Patrick Bryant, John Alison and Mikael Kuusela.