Generative machine learning methods for LHC applications

Anja Butter
Heidelberg University, Germany
Wednesday, January 22, 2020
11:00 am
NS2 1201
Event generation for the LHC can be supplemented by generative adversarial networks, which generate physical events and avoid highly inefficient event unweighting. For top pair production we show how such a network describes intermediate on-shell particles, phase space boundaries, and tails of distributions. In particular, we introduce the maximum mean discrepancy to resolve sharp local features. The generative network can be extended to perform addition and subtraction of event samples, a common problem in LHC simulations. We show how generative adversarial networks can produce new event samples with a phase space distribution corresponding to added or subtracted input samples. We illustrate its performance for the subtraction of the photon continuum from the complete Drell-Yan process.
Julian Heeck