Speaker: Theo Heimel, Heidelberg University
Abstract: QCD splittings are among the most fundamental theory concepts at the LHC. In this talk, I will present how conditional invertible neural networks, a realization of normalizing flows, can be used to extract posterior distributions for QCD theory parameters from low-level jet observables. This approach expands the LEP measurements of QCD Casimirs to a systematic test of QCD properties. Starting with jets from a toy parton shower generator, I will discuss the effect of the full shower, hadronization, and detector effects.